PART I
Philosophy of Information
Chapter 1
Subject Is No Object: Complementary Basis of Information
Piotr Bołtuć
University of Illinois Springfield
One University Plaza, UHB 3030
Springfield, IL 62703, USA
Warsaw School of Economics, Poland
[email protected] Subject is not an object, at all. This is true as long as we try to analytically purify those notions. We move the objects to one side of the equation, irrespective of their ontological status: material, mathematical, phenomenal objects, even qualia. What’s left on the other side is epistemicity, which is the subject side of the atomic subject–object relationship. On the other side is the theory of potential objects. Only interaction between subject and object results in ontological reality; it allows for information as their shaped relationship. I build very little here, while standing on the shoulders of somewhat unexpected giants: We enjoy early Fichte and Husserl-style pure subject (contra Kant’s transcendental subject busy with categorizing things); Leibniz’s argument that unknowable universes cannot exist segues into the ontological condition of epistemicity; I follow Russell 1921, making a complementary scaffolding that starts by juxtaposing first- and third-person epistemologies; I follow Plato, Peirce, and Burgin in exploring triadic structures that transcend the basis of subject–object complementarity. The non-reducible first-person subject is shown as not merely epiphenomenal. I spend some time demonstrating how privileged access may no longer be quite privileged (Gallant’s experiment) and how Jackson’s Mary’s problem consists in her lacking the word-to-phenomenal-content link in her cognitive architecture. The main point of this part of the argument is to disperse the mist of qualia that hazes over the non-reductive subject. This exposes information as always already information for an epistemic subject, for some general consciousness.
1.Easing In
Das Subjekt ist kein Objekt. The most reduced notion of the subject is not any object. It is the locus of epistemicity. Such an object, apart from any subject, is mere potentiality. Leibniz was right in that the possible world, which is not causally connected with our world, cannot exist, since we would have no reason to assert its existence. Assertability produces an epistemic condition of any ontology. Also, epistemology without ontology is empty. Even subjects of phenomenal experience are subjects nevertheless — it is impossible to be epistemically aware of nothing.
The above is true of the most reduced notion of subject (Fichte’s or Husserl’s) and object (Russell’s atomic object). A similar structure replicates at the level of building consistent conceptual frameworks, or basic philosophies.
Non-reductive epistemic subject is impossible to establish within ontology. Epistemic subject boils down to an epistemic grasp of objects, but such epistemicity itself can never be grasped. Likewise, ontological entities are impossible to establish within first-person epistemology. Here comes an unexpected analogy between the status of subject-perspective and object-perspective: Within ontology, pure subject seems epiphenomenal — it changes nothing in the causal structure of the world, and postulating it is superfluous within ontic framework. Surprisingly, within consistent firstperson epistemology, those are ontological objects that seem epiphenomenally unnecessary (Berkeley saw this, but failed to provide its non-religious account.) Thus, in a purely epistemic (phenomenal) frame of mind, objects with ontologies that go beyond phenomenal experience become . . . well . . . strangely epiphenomenal. They change noting in the workings of such a world, and the postulate of their existence turns out to be superfluous. Those perspectives function complementarily.
If any ontology is impossible without relation to the subject, then this is even more so specifically for ontology of information. Exemplis gratias, Floridi’s theory of information is built as an ontology. He tries to provide its epistemological grounding with his Kantian starting point, but replaces it with realism along the way.a If we are to take Leibnizian condition of epistemicity seriously, Floridi’s theory of information would be better viewed as an epistemology; yet this would be a different mode of his theory than the one he emphasizes.
Burgin’s General Theory of Information provides a conceptual framework within which this article can be seen as a study on some of the basic philosophical grounding issues in ontology of information. In particular, ontology must always be grounded in epistemicity and, vice versa, epistemology must always be defined in relation to some objects. My focus on epistemicity could easily be misunderstood as anthropologization and subjectivisation of information theory. Yet, epistemicity does not lead to anthropocentrism or subjectivism: Today’s theory of Artificial General Intelligence is an empirically grounded view that presumes continuity between human, animal and artificial cognitive architectures. In particular, Goertzel’s cognitive synergy approach [Goertzel] shows how epistemicity does not have to be human-centric.
This has been a broad-brush roadmap; now we are ready for a proper introduction.
2.Introduction
In this chapter, we tackle one of the deepest issues in philosophy: the relationship between the two sides of the human experience. Some call it subjective and objective, others mind and body, or self and the world, or awareness and its object, still others the epistemic and ontological viewpoint. We argue for the following methodological approach: To begin with, one should look for the most basic, reduced conceptual framework. Only clarity at the starting level gives us the sharp distinctions required to go beyond those basic dichotomies.
It makes sense to distinguish the subject from all the objects. The objects may have different ontologies, some are material things outside of our bodies, others are objects of thought (thus, confabulations of our minds), still other objects may be theoretical generalizations built upon directly observable ones. Many philosophers think that the subject (self) has a special status among the objects. We argue that such status is very special indeed, since — in its most straightforward sense — the subject is not an object at all! It can be viewed, metaphorically, as a mirror that reflects the objects of all kinds, while not being any of them; but this is merely a metaphor. The subject, in its most reduced sense, is not an ontological object. It is nothing but pure potential for epistemicity. In its less reduced sense, it appears as an object to others, and to oneself — but this is more of its outer shell, which in fact is an object. Objects that are not subjects may interact in the universe of possibilities, but they are merely potential; they can be said to actually exist only as, and as soon as, they present themselves to a consciousness (thus, an epistemic subject). In the ontological narrative, objects are possible if and only if such (subject–object) interaction occurs with a special higher-level-object — the one that is characterized by direct epistemicity. This point requires an explanation: atomic objects do not have epistemicity — since pure epistemicity is not an object. Yet, objects form more complex structures, high-level monads of sorts. Some such higher order structures have the feature of first-person epistemicity. This is the subject viewed from the outside, which is objectified as an it. The relation of epistemic objects is much like Leibnitz’s conscious monad among the unconscious ones.
This point can fully be appreciated only in the complementary framework, in which we see the first-person (epistemic) perspective as one aspect of the world accessible to human cognition, and the third-person (ontological) perspective as the other aspect. None of those aspects has priority — just like in the old-style wave–corpuscular theory of light, each of the complementary descriptions has some unique advantages over the other. There also has been no known more general framework allowing one to retranslate the two complementary frames. It is possible that, one day, humans may discover that we have more such cognitive viewpoints (the second-person relationship is a strong contender in philosophical anthropology and ethics), or we may even attain an overarching meta-theory. But the complementary framework, of the pure subject of epistemicity and the pure ontological object, seems to be the most analytically reduced, and hence conceptually most basic, perspective.
The complementary framework looks like the right way to gain more than we would be able to endorse within the sole theory, either pure phenomenalism (on the subjective side) or reductive physicalism (on the objective side). This way we do not have to cut human experience to fit within the Procrustean bad of one limited framework, whether phenomenalism or a sort of reism. We work with the most reduced notion of pure subject, based on Husserl’s view of the passive transcendental subject, which may rely on the pure transcendental subject from Fichte’s second introduction to the theory of knowledge (which is only a distant relative to Kant’s transcendental subject active in the process of constituting the world). Importantly, we also endorse Leibnitz’s point that possible worlds that do not interact with our known universe cannot exist — this is just because we would not have any reasons to make judgments about them. This simple point establishes strong (sine qua non) relevance of epistemology for ontology. Every ontology seems to need epistemicity to acquire its fully blown ontological status — and not just a status as potentiality.
Information-based ontology is an ontology after all; so, Leibniz’s argument about unknowable possible worlds applies. We try to demonstrate that the epic attempt of Floridi to have ontology (and ethics) based on information reveals the edge beyond which ontology always already turns into (or, at least turns towards) epistemology. Critiques by Beni (the theory of information is not grounded in ontology but epistemology [Beni]), Barker (the too much information approach [Barker]) and Fultot (the information as a servant to entropy approach [Fultot]) may be viewed as one way of exposing the need to center informational ontology around epistemicity. The epistemic framework needs to be added to the supposed infinity of subject-less information advocated by Floridi. If so, epistemic subject is sine qua non of any ontology, and thus sine qua non of ontology based on information.
After multiple introductions, it is time to move to the main body of the chapter, and to move a bit more slowly at that.
3.Subject and Object
Das Subjekt ist kein Objekt — this German sentence conveys the idea that “the subject does not belong to the set of all and only objects”; it does so a bit more clearly than the English “A subject is not an object.” The English version may easily be misinterpreted as a trivial idea, analogous to the sentence “an apple is not an orange”. Our point is that epistemic subject is not an ontological object at all, it does not directly belong to the class of objects. It belongs to the class of objects only indirectly. The subject presents itself through the manner in which it is to the other observers (and also to oneself, if put in some sort of metaphysical mirror that places it among the objects in the world). This happens if one engages in self-reflection, proprioception or, say, watches oneself in a mirror. The subject under such description manifests itself through objects, a bit like a ray of light, or a shadow. In most situations, people talk of subjects in a broader way, as complex objects with some epistemic features, such as human beings, animals, maybe societies or robots. But the notion used in those contexts is not quite the notion of the subject as such. My regulative definition of the notion of the subject tries to refer to the most distilled notion of the subject. This notion needs to be distinguished from a broader definition of subject as an object characterized by some features of epistemicity (such as perception or agency). My goal is not to abandon the other, more relevant in real life, uses of the concept of subject, but to purify the gist of the notion of the epistemic subject, so as to have clarity of what it is. This is accomplished by defining a pure epistemic subject.
The most reduced definition of pure epistemic subject leaves such subject with no direct predicative features. To restate it, we can predicate about it, the way we do in the present sentence, but we cannot predicate of it, in the narrow sense of providing a direct description. By predicating about the epistemic subject, we use a meta-level of reference — the difference is, roughly, between that of a person people talk about at a party versus characterizing someone present at the party, known to us in person (speaking of him or her). Every attempt to predicate directly of the pure epistemic subject would make us miss the point of such attempted reference. This is because we would be describing some object, no doubt related to the epistemic subject, but not the subject — not the features that make it the subject. The sole grasp of pure subject is from the inside. From the outside it can be viewed, at most, as the empty space, so to say, that remains after we put all the objects to one side.
Why would we attempt to move all predicative features to one side? It is primarily for the sake of clarity. Specific features of some object do not overshadow what one is looking for when searching for pure subject-hood. In algebra, it helps to move all variables to one side of the equation; also in simple chemistry and other natural sciences, one defines the unknown substance by leaving it alone on one side of the equation. There is much confusion among people who try to grasp the epistemic subject by identifying it with a certain set of advanced features, such as memory, secondary qualities (Locke) or the feel of “what it is like” (Nagel). Those features are just intuition pumps (Dennett), quite misleading under literal interpretation.
In his early works (original and best), Thomas Nagel [Nagel, 1979; 1987] identified subjectivity as the view from nowhere. Bertrand Russell — in the early version of his neutral monism [Russell, 1921], which seems so much better than Spionza’s pantheism, and Russell’s own later and more discussed neutralism [Russell, 1927] — juxtaposed the subjective perspective to the view focused on objects. He established a complementary framework where two descript...