S.L. Katrechko

Imagination as fundamental faculty of consciousness

These are the base hypotheses of our paper (research project):

1.   Faculty of imagination is basic mechanism of conscious activity, which determines other human faculties. Its main appointment is to form images, or totalities.

2.   Philosophical activity is an interesting manifestation of this faculty. Philosophy itself can be represented as a kind of «work» with special metaphysical objects; and its history  — as a consecutive change of «paradigmal» totalities.

3.   The development of algorithmic imagination mechanisms (so-called fantasy mechanism and transition on meta-level) brings up to appearance of qualitative new type of «artificial intelligence» systems — intellectual systems.

There are 3 comparatively independent objectives (problems) within the project:

·      (a) A development of hypothesis 1 based on Kant’s conception of knowledge;

·      (b) A development of the line hyp. 1 — hyp. 2;

·      (c) A development of the line hyp. 1 — hyp. 3.

Kant’s conception of cognitive process will be modified to solve the problem a [1, 2, 3]. The main idea is that the base of knowledge is a synthesis of images — totalities carried on out by the faculty of imagination. It’s this fact predetermines specific character of human knowledge. The synthetic activity of imagination reveals itself on both levels: sensibility and understanding (mind). Note, that our interpretation of «image» as «totality» expands the term comprehension, because totality includes sensible perception («pictures») — traditional image conception — as well as intelligible (supersensible) concept. Firstly, Kant’s a priory sensible and intelligible forms are the «pure» images received with the help of transcendental faculty of imagination. Secondly, images as products of the empirical faculty of imagination make up a basement for not only the sense-perception, but for the intelligible (rational, theoretical) ones. These are the schemes (Kant’s «schematism of the pure concepts of understanding») of certain kinds of practical activity (M. Vartofsky).

We offer the following conception of knowledge activity. Sensibility delivers us only a primary (raw) empirical material. Faculty of imagination («intellectual intuition» of Leibniz, «intuition apprehension» of Kant, «eidetic intuition» of Husserl) transforms it into image (totality). For example, a human eye perceives a picture not only as a set of colored stains and strokes (although such perception is also a result of imagination apprehension (synthesis) at a lower level), but as a landscape or/and depictions, such as «a knight», «horse» etc. The same occur, when perceiving with an ear a combination of music tones, we hear melody in the whole (totality of melody). After that faculty of understanding (mind) begins its work. The primary — synthetic — intelligible activity includes, firstly, a generation of concept based on images, and, secondary, further uniting of the concept into sentences, witch normal structure is «A – B» (Aristotle). It’s a sentence that plays role of the elementary language cell (L. Wittgenstein). And the secondary — logic-analytical — mind activity takes place in the language itself and comprises: (1) construction reasoning and (2) development and enrichment of language.

The original conception of philosophical activity as metaphysics will be developed so that to find the solution of the chief problem b [4, 5, 6]. The essence of the approach is that the philosophy is a particular kind of «work» with special «metaphysical objects», or «totalities». Philosophy differs both from science, which research and deals with «exterior» objects («physical» objects par excellence), and from art, which investigates «objects» of the «interior world» of a person. Philosophy is in some way a rationally discursive activity (the sphere of rationally-discursive knowledge), which differs both from science and from art (for example, literature).

What are the «totalities»? Let's clarify it by considering one of them as an example. One of such objects is the «world» («universe»), and, in this sense, we often say that philosophy is the «weltanschauung» (world-outlook). Certainly, the world as some set of things can be studied by «physics» (sciences in a broad sense), but in this case a researcher can miss the point that the world is a totality, not just a simple set of things. Philosophy, being distinct from scientific study, takes its objects as «totalities», which the subject of knowledge cannot study as ordinary objects, because the subject-embracing nature of these «totalities» excludes any standard scientific approach to knowledge in principle. The human being, as the subject of research, is always «inside» these «objects» and never can «jump out» of them. Other examples of objects mentioned above («metaphysical objects», or «totalities» have been obtained with of imagination faculty) are other Kant’s «idea of reason» («Ego» and «God»), Plato’s «idea», Descartes’ «innate ideas» and Deleuze’s «concept». Objects of this type (with some reservations) are many human being-related phenomena, such as «love», «virtue», «conscience», «courage», «bravery», «understanding» and so on. «Metaphysical objects» cannot be given in the frameworks of «physical» experience: they are our mental constructs. «World», in contrast to any final collection of things, is not given to us, like a mental construct of a set (or constellation — N. Goodman’s example), which a mathematician (human being) creates with available elements, is not given to him. Specifying the status of these objects, we can say that the contents of philosophical activity are metaphysical-transcendental objects, i.e. objects with the boundary status of their existence. They are, for example, neither subjective nor objective, neither conscious nor material. In some sense, they are analogues of Husserl`s phenomenon.

The postulation of special «metaphysical objects» of philosophical work predetermines specificity of the language and reasoning («logic») of philosophical discourse, methods of philosophizing. Let's formulate main hypothesizes we offer for solution of the raised task. Firstly, the language of a philosophical investigation is the special symbolic language («symbols» are names (signs) of «metaphysical objects»). The language of philosophy is the language of «symbols»; «symbols» are the «house» of philosophy. Thus, as distinct from usual languages, it has not three-level semantics of the usual language: a level of syntactic words, a level of senses and a level of meanings (references) (G. Frege), but two-level (non-referential) semantics: a level of syntactic words and a level of senses. Symbolical language, on the one hand, limits possibilities of philosophical discourse (as an appeal to a referent, experience is impossible), but, on the other hand, expands possibilities of the latter, because philosophy speaks not only about the real «earthly» world, but also about any «possible world» (Leibniz). Secondly, the «metaphysical» («symbolical») nature of the categorical philosophical apparatus predetermines also the specificity of philosophical reasoning. The structure of reasoning about «pure being» (or «conscience») as «mental construct» differs in essence from the reasoning about, say, a tree or other objects of the «earthly» («physical») world. The characteristic of «metaphysical objects» as «mental constructs» allows raising the question: which logic describes adequately the introduced «metaphysical objects» and how much the application of common «down-to-earth» Frege—Russell logic with the principles of non-contradiction and excluded third is correct ? And more than that, it is possible that another logic, distinct from usual, «earthly» logic, operates in the sphere of speculative objects. In particular, according to the opinion of Russian philosopher and logic N. Vasiljev, validity of the logic principle of excluded third has a purely «earthly» origin and is connected with primacy, after Aristotle, of individual existing things. If we, following Plato, consider classes of subjects as primary (for instance, subject is a class of tables), the principle of excluded third will be incorrect, since the table as one of the set of subjects of the class of tables can be simultaneously both white and non-white, for instance, green or red (of course, here it is necessary to realize, that the opposite properties are attributed to the different individuals of the primary Plato’s «subject» — different things of the same class). The transfer of the concentration from single objects to classes of objects has allowed Vasiljev to formulate «imagined (paraconsistent) logic» with the principle of «excluded fourth. By its logical firmness this logic concedes no point to our habitual logic with the principle of excluded third, but expands possibilities of our thinking. In this connection it should be noted that the criticism of the principle of the excluded third by Vasiljev differs in it motivation from intuitionism. Outlined at the beginning of the 20th century, Vasiljev’s approach has predetermined the development of paraconsistent (A. I. Arruda — N. Da Costa) and truth (Von Wrigth) logic.

Besides, the conception of the philosophy as the work with «metaphysical objects» allows to implement the following original interpretation (logical reconstruction) of the accumulated historico‑philosophical material. Its essence is following. Any philosophical system can be represented as a certain structure (system) of basic «totalities» or «symbols», which forms its conceptual framework. Accordingly, the history of West European philosophy is a consecutive change of such «symbolic» systems. Three replacing each other «episthemes» (Foucault) or «paradigms» (Kuhn) of philosophizing can be distinguished in these systems development: ancient, classical (the New Times) and modern. The distinction of these three large meta-systems of philosophizing is caused by the change of the «objects» of the philosophical research during the transition from one «paradigm» to another. The attention of the ancient thinkers was focused on such «totality» as COSMOS (WORLD in our discourse) and the basic problem of philosophical research could be formulated like this: how COSMOS (in contrast to CHAOS) was possible, i.e. what «entropyless» (meta‑physical) mechanisms (means) of conversation CHAOS into COSMOS operated there. Various ancient philosophical systems answered the question differently. For example, according to Pythagorean school a basis of regulation and harmonicity of COSMOS was a Number. Parmenid postulated «Uniform», or «Being», as an «entropyless» principle (source) of COSMOS. The transition from the ancient «paradigm» to New Times (classical) one is connected with the «splitting» of the one object of philosophical research of COSMOS into three «ideas of reason» (Kant): «Ego» («soul») — «World» — «God». A new interesting object of investigation appears in a field of vision — «subject» (in contrast to the «object»; «subject versus object»), for instance, Descartes' «cogito» (the New Times) or «God» (according to the medieval theology). Correspondingly each philosophical system of this period, as construction of «metaphysical» object «Ego» — «World» — «God», implicitly solves the problem of correlation among these «totalities», which of them is «primary» and determines the other components of the system. The transition to the modern type of the philosophical discourse is predetermined by two «events». The first of them is «Death of God» (Nietzsche); the second is «Death of the subject» (structuralism, postmodernism). So, on the one hand, it makes three «ideas of reason» as if «clap» into one — WORLD, and, on the other hand, instead of the ancient united COSMOS a number of pluralistic «worlds» of different philosophical systems are appearing. Schopenhauer considers «world» to be «Will» and «Idea»; for Marx — «world» is, first of all, a system of «social relations». Husserl, instead of the New Times contraposition «subject vs. object», investigates so-called «phenomena», having elaborated the conception of the phenomenal («vital») world; structuralists suggest that we should consider «world» to be an aggregate of «language structures». In this variety of the pluralistic approaches similarity concludes in attempts of thinkers within the framework of modern «paradigm» of philosophizing to explain the «world» as a certain primary «entity», i.e. to explain it out of itself avoiding too «strong» assumptions (hypotheses).

As it was said above the development of the c in the project is aimed at the perspective. In contrast to a and b solutions it’s supposed to define more precisely the offered hypotheses about the main mechanism of imagination faculty. For this purpose the available investigations in theory and methodology and practical systems in the «artificial intelligence» field were analyzed by author. On the other hand, it must be noted that the author of the project has already turned to the theme «From logic calculus to intellectual systems». It was studied in a certain way in the works [7, 8, 9] and the connection ac imparts some new seeing to this task. The development of a gives basic methodological means for construction the new type of intellectual systems. The well-known metaphor «right mind» versus «left mind» is a general background of the problem c. But in this case it’s considerably reinterpreted and specified: activity of «right mind» stands for the imagination faculty. It allows to propose an original methodological approach to creation of so-called intellectual systems. The statement of a problem connected with algorithmic modeling of a faculty of imagination in «artificial intelligence» is absolutely new by itself.

There are two supplementary base mechanisms form a faculty of imagination: a fantasy mechanism and transition on metalevel. A fantasy mechanism makes it possible to «complete» a partial «picture» to the whole one (image, totality). It can be clearly demonstrated by the perception 3—dimensional objects, which (it is obviously) are given as a whole in no sense perception. The essence of this mechanism is extrapolation given in perception. A mechanism of the transition on metalevel — the scale-measuring variation — allows to design totalities by changing (moving away) an observation point. This, for instance, was shown in the phenomenon mention above (see a): the perception of a picture (music melody) by a person isn’t only the grasping of colored spots (separate tunes), but the apprehension of a picture (melody) in the whole. More exactly, a human being in this case (perceiving a picture) can occupy 3 positions (points of view): 1 — apprehension of only colored spots and strokes (position of a «nearest» observer); 2 — apprehension of «images» of a picture (position of a «middle» observer); 3 — apprehension picture itself as a picture (position of a «long sight» observer).

In accordance to it two ideas of modeling faculty of imagination within the framework of the proof‑search theory, based on transition from logical calculi to metacalculi — calculi of proof‑search [7, 8], can be postulated. The first one can be called method of dummy variables (metavariables). The expansion of expressive opportunities of calculus in it is attained by the metalevel means: metavariables, metaconcepts etc. Its function consists in fixing (complete or incomplete at the certain stage of a derivation) information about a possible structure of the derivable formula, and in consequent, in obtaining an additional necessary (complete) information, replacement metavariables on variables or formulas of initial calculus. The efficiency of such approach is demonstrated by means of so-called FOL—systems (R. Weyhrauch, [10]). The second idea can be named method of «global processing of information». In this case there is a possibility to take into account structural information of logical formulas useful for proof‑search. From this point of view, the approach to a construction of algorithms for automated reasoning, so called inverse method proposed by Sergey Maslov, is interesting [11]. Inverse method’s calculi are metacalculi, in which it is possible to express structural information about «gearing» of the variables of the formula. Using of this information allows to hope for a construction of effective non-exponential (e.g. propositional) algorithms. It is possible to introduce a modification of Maslov’s inverse method, which I termed calculus of numbers with index [12]. I regard this calculus as universal proof-search calculus, in which it is possible to simulate mechanism of faculty imagination. Above all the mechanism of «creative mistake» in so-called iterative system of automated deduction of the theorem [13] is a development of great interest, because it is closely connected with a fantasy mechanism described above.

References:

1.    Katrechko S.L. Refleksivnyj podhod k issledovaniiu fenomena soznaniia //Gumanitarnaia nauka v Rossii: sorosovskie laureaty (filosofiia). — Moscow, 1996 (In Russian)

2.    Katrechko S.L. Znanie kak soznatel'nyj fenomen //Chto znachit znat'? Moscow, «Centr gumanitarnyh issledovanijy», 1999 (In Russian)

3.    Katretchko S.L. O tvorcheskoj sposobnosti voobrazheniia //Tezisy dokladov mezhdunarodnogo seminara lAntropologiia s sovremennoj tochki zreniiay. Kaliningrad, Kaliningr. u-nt, 1998 (+ enlarged new version: Katretchko S.L. Kak vozmozhna tvorcheskaia sposobnost’ voobrazheniia: www.atimidus.ru) (In Russian)

4.    Katretchko S.L. Philosophy as Metaphysics //20th World Congress of Philosophy, 1998)  (see 2-d online version).

5.    Katrechko S.L. Filosofiia kak metafizika (programma kursa). Moscow, Izd-vo: Universitet RAO, 1997 (In Russian)

6.    Katrechko S.L. Moe ponimanie filosofii (vvodnaia lekciia k obshchemu kursu filosoii) //Internet—server «Philosophy in Russia»). — Moscow, 1998 (In Russian)

7.    Katretchko S.L. Between Logic and Heuristic //20th World Congress of Philosophy, 1998.

8.    Katretchko S.L. From logic to Proof-Search Theory //Proceedings XI LMPS’99 (XI-th LMPS). — Crakow, 1999.

9.    Katrechko S.L. Ot logicheskih ischislenij k intellektual'nym sistemam: vvedenie v teoriiu poiska vyvoda (programma kursa). Moscow, Izd-vo MSU, 1994 (In Russian)

10. Weyhrauch R.W. Prolegomena to a theory of mechanized formal reasoning //Artificial Intelligence, 1980, Vol. 13. P.133 — 170 (Aiello L., Levi G. The uses metaknowledge in AI systems //Proc ECAI, pp. 315 — 324, 705 — 717, 1984).

11.Maslov S.Yu. The theory of deductive systems and its applications. Moscow, Radio i Sviaz, 1986 (in Russian); [English translation MIT Press 1987].

12. Katrechko S.L. Obratnyj metod S.Iu. Maslova //Logika i komp'iuter 2: logicheskie iazyki, soderzhatel'nye rassuzhdeniia i metody poiska dokazatel'stva. — Moscow, Nauka. 1995 (In Russian)

13. Voprosy kibernetiki (problemy sokraschenija perebora). M., 1987. Vyp.131 (In Russian)