Georgi Lazarov
Draft version September 2003
Call
for Comments
Any recommendations, criticism, and positions on the
ideas presented in this essay are welcome! Send your commentary to e-mail: aesthesionomy@yahoo.com
Keywords: phenomenal consciousness, materialism, explanatory gap, functionalism, representationalism, aesthesionomy.
The topic of the essay is the “explanatory gap” between, on one side, descriptions of conscious states from 1st person perspective, termed as phenomenal (P-) consciousness; and on the other side, the descriptions of conscious states in representational theories of mind, from 3rd person perspective, termed as access (A-) consciousness. The main source of the explanatory gap between P-consciousness and A-consciousness is the methodology of functionalism, accepted in almost contemporary representational theories.
I argue for the following: (1) The principles of materialist ontology, accepted in representational theories of mind, are true. But materialism is not necessary linked with functionalist research methods. The only reasonable consequence from the existence of the “explanatory gap” is that functionalism should be replaced with more adequate methodological approaches. (2) One such non-functionalist approach, termed as “aesthesionomic approach”, is proposed. The central idea is that phenomenal conscious states should be first described in objective terms (from 3rd person perspective). Only on base of such a preliminary objective description of phenomenal conscious states, the description of representational mechanisms (able to realize these states) will not give rise to an “explanatory gap”.
Existence of consciousness is nowadays often characterized as an epistemological mystery in the field of materialism. Epistemological mystery is problem, in which we have no idea what could prove to be a solution, or what kind of an investigation is necessary in order to find a solution.
Parallel with materialism, new non-materialist ontological theories, aimed to ‘fit’ the data of contemporary science, are developed. Chalmers (2002) claims that at least three types of non-materialist ontological theories are actual alternatives to materialism:
– interactive dualism,
– epiphenomenalism, and
– panprotopsychism, i.e. the view that phenomenal or protophenomenal properties are located at the fundamental level of physical reality.
He concludes:
It is often held that even though it is
hard to see how materialism could be true, materialism must be true, since the
alternatives are unacceptable. As I see it, there are at least three prima
facie acceptable alternatives to materialism on the table, each of which is
compatible with a broadly naturalistic (even if not materialistic) worldview,
and none of which has fatal problems.
The non-materialist theories of mind have no problem with the explanation of consciousness. Why do we have to select as an ontological research framework the materialist view, the only one in whose scope consciousness seems to be a mystery?
Physics provides explicitly only one constraint over the ontological theories, namely the conservative laws. This argument is restricted only against the interactive dualism. And even the interactive dualism succeed to “fit” this constraint, by claiming that an immaterial mind could activate some neural response by “increasing the quantum probability” of some biochemical processes in the cerebral cortex “without interfering with the conservative laws of physics” (Eccles, 1994, p.11).
The contemporary opponents of materialism claim that scientific evidences against non-materialist ontological theories do not exist at all. Eccles (1994, p. ix) told sarcastically: “Materialists remain as dominant as ever because they are devotees of a dogmatic belief system, which holds them with a religious-like orthodoxy”.
My replay is that materialist ontology is based not only on data from neurophysiology, psychology and conservative laws of physics; materialism is a derivative of the entire system of contemporary science. We know from physics and chemistry that atoms and non-organic molecules, which build our bodies, have no mental properties. Cosmology presents us a history of the universe from the Big Bang till the present, and in this history there is no place for immaterial souls. In order to “add” immaterial souls to human brains (or protophenomenal properties to the fundament of the universe), one has first to change contemporary cosmology.
The general ontological debate on the nature of mind is not “closed”; but I claim that this general ontological debate must be separated from the debate about the consciousness in the scope of materialism, because the arguments used in the two debates are radically different.
In this essay I take as a given that human conscious experiences do arise (somehow) from physical and chemical processes in the human brain, and I consider only the further question: How and why does it happen?
According to agnostic materialism (called also “mysterialism”), there are good reasons to accept materialist ontology, but existence of consciousness will be forever a mystery. Human cognitive abilities possess some inherited limits making the scientific explanation of the mind impossible. Colin McGinn (1989/1997) introduced the term “cognitive closure”
A type of mind M is cognitively closed
with respect to a property P (or theory T) if and only if the concept-forming
procedures at M’s disposal cannot extend to grasp of P (or an understanding of
T). …
What is closed to the mind of a rat may be open to the mind of a monkey, and what is open to us may be closed to the monkey. Representational power is not all or nothing. Minds are biological products like bodies, and like bodies they come in different shapes and sizes more or less capacious; more or less suited to certain cognitive tasks.(p. 529-530)
We have to agree with McGinn that any mind, at any time is cognitively closed with respect to some aspects of reality. But the agnostic conclusion, made further by McGinn about the theories of mind, is wrong. The limits of concept-forming procedures are historically determinate. Some other problems were earlier also considered as unsolvable in field of materialism, but however people succeed in solving them. History of science contains ‘know-how’ for solving epistemological mysteries in materialist ontology, and we can use it by solving the problem of consciousness.
Let us consider briefly the problem of origin of life. We can point introspectively two features of living bodies, which had made the problem to appear as unsolvable until the end of XIX century.
The first feature is the complexity of living bodies. While trying to explain the “origin of life”, philosophers and biologists tried to explain how non-living particles could be linked together by some natural process in a complex body as a horse or a bird. Defenders of creationism claimed that such ‘natural process’ is impossible. The Development of the theory for the evolution of the species and the theory of cell building of living organism changed the view on what has to be explained in theories of the origin of life. The first few non-mysterious materialist theories of the origin of life, developed at the beginning of the XX century, were theories about the origin of simple one-cell organisms.
The second feature, which made the problem to be viewed as unsolvable, is the “teleological organization” of all processes in the living body. Immanuel Kant claimed in the "Critique of Teleological Judgement" ( Kant,1790 /1972, §77) that no human mind, neither any other possible finite mind, (i.e. anyone except Got) would be able to explain with physical causality the activity of a living body, even a simple one, like a grass. His argument was purely epistemological - physical causality gives explanations from the present to the future, but activity of any living body is dependent on the result in the future. It was not until the 1930s that biology accepted the simple concept of feedback control as sufficient for the elimination of the epistemological mystery of teleological organization of processes in living bodies.
Let us make a conclusion from this story. The epistemological mysteries were caused by some specific historical constraints on scientific knowledge. But scientific knowledge could be changed, and was changed in course of investigation; respectively the problem was demistificated. What was in 1800s beyond the cognitive limits, is today within the limits.
This conclusion seems trivial in retrospection, but it is very important in considering the problem of consciousness. Kant’s error by explanation of life was that he extrapolated over the future the historical constrains of science. Agnostic materialism makes the same error.
The plan of the essay is the following.
In section 2 I describe a version of materialism, which incorporates the ideas of (a) emergentism – the view that the mind is a system of emergent processes which supervene on neuronal processes in the brain.; and (b) representationalism - the view that the mind is a system creating representations of the environment, and thereby using these representations for control of behavior. I term this view “materialist emergent representationalism” (MER). I present MER schematically, in form of a small set of ontological and epistemological principles, without giving arguments for them. But the principles accepted in MER are not new, and for any of them could be found many arguments, given by many authors.
In section 3 I consider the explanatory gap between phenomenal psychology and the representational theories of consciousness. I maintain that the source of the gap is the methodology of functionalism, accepted in almost contemporary representational theories. But the materialist ontology is not necessary linked with functionalist research methods. Therefore, the only reasonable conclusion is that functionalism should be replaced with more adequate methodological approaches.
In sections 4 I present a non-functionalist approach, termed as “aesthesionomic approach”. The central idea is that phenomenal conscious states should be first described in objective terms (from 3rd person perspective). Only on base of such a preliminary objective description of phenomenal conscious states, the description of representational mechanisms (able to realize these states) will not give rise to an “explanatory gap”.
The aesthesionomic approach is not ontologically neutral. The principles of MER constitute its ontological framework.
In section 5 I analyze the relationship between the aesthesionomic approach and the various representational theories of consciousness is considered. The aesthesionomic approach is compatible with most of the contemporary representational theories of consciousness, if they are interpreted according the principles of materialist emergent representationalism.
Materialist emergent representationalism (MER) is termed a view on the nature of mind, based on the following ontological principles:
(MER 1) The general principle of materialism: The particles on the fundamental level in universe (sub-atom particles, atoms, and non-organic molecules) have no mental properties.
(MER 2)The principle of emergent representationalism: The mind is a system of emergent states and processes in the brain; i.e. mental states supervene on neuronal processes in brain, but can not be identified with neurophysiological processes, neither can be explained in terms of neurophysiology. The mind is an emergent system, which creates representations of environment and internal states of the living organism, and uses these representations for control of behavior.
(MER 3) The principle of the dual-coding organization of the human mind: Human representational system consists of two representational subsystems: depictive (nonverbal) representational subsystem and verbal representational subsystem. The depictive representational subsystem stems from an earlier evolutionary stage as verbal representation. It can exist within the absence of abilities for verbal representations. The existence of verbal representations presupposes the existence of depictive representations.
Terminological note: Within the verbal representational subsystem the representations are termed as propositional representations, thoughts or concepts. Within the depictive representational subsystem the representations are termed as depictive representations or mental images.
(MER 4) The anti-sensibilist principle Sensibility (the ‘feel’ ability) is not a fundamental property of the mind. The mind originates in biological evolution in the form of representational systems by which all mental states are unconscious states. Sensibility emerged as a subsequent mental ability. Conscious processes supervene on the primary (non-conscious) representational processes.
(MER 5) The principle of the multi-level organization of the science of mind: The entire system of scientific knowledge about the mental activity of the brain consists of several hierarchically subordinated levels of explanation. The processes on any higher level emerge from (supervene on) the processes at the lower level.
We have to distinguish at least five main levels of the explanation of the mind/brain:
L4. Level of conscious mind
L3. Level of primary (non-conscious)
representational states
L2. Neurophysiological level
L1. Biochemical /cytological level
L0. Fundamental
physical level
MER, as present above, has been in existence before the middle of XX century. The revolution in the science of the mind based on MER, during the second half of the XX century is linked with the development of information-processing theories of intentionality (IPTI). IPTI accepted some non-traditional ideas about the relationships between, on one side, intentionality and consciousness; and on the other side, intentionality and neuronal processes.
The kind of materialism, dominating till the middle of XX century, is the sensibilist materialism. This view comes from the French materialism of the XVIII century. According to the sensibilist materialism, the initial form of the mind is sentience (where sentience is the ability for simple conscious experiences). Respectively, any form of the mind has both intentional and phenomenal properties; any mind is a conscious mind.
According MER, in contrast with sensibilist materialism, the initial form of the mind is the non-conscious mind (primary representational systems) which has no kind of conscious experience. Respectively, fundamental theories of intentionality (theory of representational properties) could be developed independently of explanation of consciousness.
The separation of theories of intentionality from theories of consciousness does not mean that one has to explain the content of the conscious mind independent of consciousness. The conscious mind is a later result of the development of the mind. It constitutes a new set of emergent properties, which need additional explanation. Criticism against theories, trying to explain the content of consciousness by ignoring the phenomenal features of the conscious states, is presented by McGinn (1991 /1997). He pointed clearly that such theories imply the claim that conscious states are epiphenomena. MER is an anti-epiphenomenalist view. So we have to agree that “we cannot hope to devise an exhaustive theory of the nature of conscious content, while remaining neutral on whether such content is conscious” (McGinn, 1991 /1997; p. 300) According to MER, fundamental theories of intentionality should not explain the content of the conscious mind, but only the intentionality of primary (non-conscious) mind.
The second key idea in IPTI is about the question:
Which aspect
of neuronal processes in the brain is essential for intentionality?
The traditional answer in materialism is that some substantial feature, specific to the brain, is necessary for intentionality. The problem of this view is, that neuronal cells have nearly the same biochemical and physiological features as any other kind of cells.
In second half of the XX century, in contrast with the traditional view, the idea, that intentionality can be explained relatively independent from the specific substantial features of neuronal cells in brain was developed. We can consider the brain as a linked set of abstract automata, described only in terms of inputs, outputs, internal states and transformation functions. Such description in terms of formal theory of automata is termed “abstract neuronal network” (ANN). The idea is that any aspect of the physiology of the brain, which is significant for intentionality, could be presented in the description of ANN. Physical and chemical properties of neuronal cells are important only as far as they realize some features of an ANN.
(MER 6) The principle of information-processing theories of intentionality: Intentionality supervenes only on such features of neurophysiological processes in the brain, which could be presented by descriptions of the brain in terms of abstract neuronal networks.
Any ANN could, in principle, be realized in physically different systems. So an ontological consequence from IPTI is:
(MER 7) The Principle of multi-realization: Two representational systems (minds) could be equivalent, but realized in physically different systems.
The principles of MER have a restrictive role by development of IPTI.
First, many claims in IPTI are based on results, deduced first in the classical computational theories. But the results from classical computational theories can not be implemented immediately in research activity of the brain. Brains are not realizations of Von Neumann machines. The results from classical computational theories can be applied nowadays by research of brain, only as far as we have theories, which can prove the validity of these results for ANN.
Second, there is a broad class of computational theories of intentionality, popular in the second half of the XX century, which is refuted in MER. Let us call this class of theories, following Georges Rey (1997), Computational-Representational Theory of Thought (CRTT). CRTT is based on the idea that all mental representations could be considered as propositions in some unified Language of Thought (LOT). Ray (1997) writes:
It presupposes to capture the attitudes
states by positing different relations born by the agent to sentences in a
special language; called “language of thought” (
According to MER, the idea of
The development of IPTI has an important influence on the general architecture of the science of mind/brain in the field of MER, presented schematically on fig 1.
|
L4. Level of conscious mind |
|
L3 Level of primary
(non-conscious) representational states L3a. Intentional level L3b. Computational
level |
|
L2. Neurophysiological level |
|
L1. Biochemical /cytological level |
|
L0. Fundamental physical level |
The level of primary (non-conscious) representational states (L3) takes central place in this system of theories. The primary representational processes, on one side, supervene on neuronal processes; on the other side, they give rise to the conscious processes.
Sublevel L3a (the intentional level) includes the information-processing theories of intentionality (IPTI).. Theories of intentionality are essentially theories of mind, and belong to psychology; even most of researchers in this area do not call themselves “psychologists”. An interesting discussion on river contemporary theories on intentionality is present in Clapin (2002)
Sublevel L3b (the computational level) includes two kinds of research: development of theories of abstract neuronal networks and the application of these theories by description of neurophysiological processes. The former kind of research belongs merely to the mathematical theory of automata. The later kind of research is often considered as part of neurosciences and is termed as computational neurophysiology.
According to MER, conscious states supervene on primary intentional states. Explanations of consciousness, based immediately on the neurophysiological or computational levels, are impossible.
Let summarize.
The term materialist
emergent representationalism (MER) is used here in order to distinguish the
described above view from alternative materialist and non-materialist
ontological views on mind. The broad terms “materialism”, “emergentism” and
“representationalism” denote some aspect of this ontological view, however, not
all views on mind, which could be characterized as “materialist” and “emergentist” and
“representationalist”, do belong to MER.
I use the cover term phenomenological psychology for any scientific research of the mind based on the 1st person (subjective) perspective.
The introspective psychology of the XIX century is the historically first form of phenomenological psychology. In the first decades of the XX century, many psychological schools – the “structural psychology”, “functional psychology”, “Gestalt-psychology” and others - refuted the introspective methodology and replaced it with other methodological principles, but they continued to be based on data from the 1st person (subjective) perspective.
In the second half of XX century, the borderlines between the psychological schools became fuzzier and phenomenological psychology could be considered as (more or less) integrated research area.
Psychological theories intended to explain the mind from the 3rd person (objective) perspective were developed only in the last forty years. This new area of psychological research was called cognitive psychology. The historical roots of cognitive psychology are in the theories of intelligence from first half of the XX century, and some versions of “purposive behaviorism”.
Cognitive psychology is not restricted to ‘cognitive processes’, as distinguished from other kinds of mental processes (‘perceptive processes’, ‘mnemonic processes’, ‘voluntary processes’ - as these terms are used in classical psychological schools). In the first monographic work, published under the title “Cognitive psychology”, Neisser (1965) marks explicitly the untraditional, extended use of the term “cognition”:
As used here, the term ‘cognition’
refers to all processes by which the sensory input is transformed, elaborated,
stored, recovered, and used. It is concerned with these processes, even when
they operate in the absence of relevant stimulation, as in images and
hallucinations. ... [G]iven such a sweeping definition,
it is apparent, that cognition is involved in everything a human being might
possibly do; that every psychological phenomenon is a cognitive phenomenon.
In cognitive psychology, like in phenomenological psychology, the subject matter of research are all mental states and processes - conscious or unconscious, voluntary or involuntary, linguistic or non-linguistic. It differs only on its goal, all mental processes to be described in objective terms, i.e. in terms defined from 3rd person perspective.
In cognitive psychology, the mind is described as a representational system (or representational mechanism). Therefore the cognitive theories are termed also as representational theories.
The most difficult problem occurring with representational theories is explanation of conscious states. In this essay I use a terminological proposal, made by Block (1995 / 1997) and accepted nowadays by many authors. We will distinguish:
- A-conscious state - states in the representational mechanism, distinguished by objective criteria, defined as conscious states in some representational theory of mind. “A-” stands for “accesses” or “availability”. The most often used objective criterion for distinguishing conscious from unconscious states in contemporary representational theories of mind is, that conscious states are directly available for global control; or equivalently, that global control process has access to them.
-
P-conscious
states – conscious experiences, as subjectively experienced, and as
described in phenomenological psychology. “P-”
stands for “phenomenal”.
Using this terminology, we can tell that if a representational theory of human mind is developed correctly, then A-conscious states must be identical with P-conscious states. The problem is that such identification is impossible by contemporary representational theories.
|
L4. Level of conscious mind |
||
|
Cognitive psychology |
The explanatory gap |
Phenomenological psychology |
|
L3 Level of primary (non-conscious) rep | ||