6 minute read

This post is a summary of Taehee Kim’s paper “Implications and Limits of Naturalizing Phenomenology.”

Before Reading

To address questions about the consciousness of humans, animals, and machines, it is crucial first to analyze the first-person experience of consciousness. Upon reflection, one can realize that most of our mental contents, such as visual and auditory perception, imagination, and temporal consciousness, are incredibly complex. For instance, consider the question, “Does a camera possess visual consciousness?” Until we press the button, the camera merely displays momentary information on a screen, which is fundamentally different from our visual perception. Our visual experience at any given moment is inherently linked to the preceding visual experience. If our current visual experience had no relation to the previous one, and if we were to experience each new scene as a fleeting, isolated event, no conscious experience could be sustained or even exist.

The analysis provided here may not be entirely rigorous. Fortunately, such a rigorous analysis of subjective experience was undertaken over a century ago—this field of study is called phenomenology. Phenomenology explores the relationships between various forms of consciousness (noesis) and the various objects they intend (noema). What insights from phenomenology can contribute to contemporary neuroscience research and theories of consciousness? These are the core questions I had in mind while reading this paper:

  1. What is the relationship between phenomenological research and a naturalistic approach? Can they be reconciled?
  2. How can phenomenological theories and neuroscientific-naturalistic theories of consciousness interact or coexist?
"Explanatory Gap problem, Naturalizing Phenomenology, Neurophenomenology and Generative Passage interpretation." Generated by Midjourney AI

1. Introduction

Chapter 1 begins with a discussion of the “Naturalizing Phenomenology (NP)” research program, a topic addressed by both cognitive scientists and philosophers. The goal of NP is to integrate phenomenological research within the explanatory framework of natural sciences. Issues related to consciousness can be broadly categorized into two major problems:

  1. The Explanatory Gap: How can the existence of qualia be explained by brain processes? _ Joseph Levine
  2. The Hard Problem: How do physical processes in the brain give rise to sensory experiences? _ David Chalmers

The key question of this paper is as follows: To assess the potential for a convergence of cognitive science and phenomenology, various discussions on NP will be reviewed, focusing particularly on Varela’s Neurophenomenology. How can the concept of Reciprocal Constraint (RC) be interpreted, and how does Neurophenomenology approach NP?

Is a convergence of cognitive science and phenomenology possible? NP is a prerequisite for this possibility.


2. Neurophenomenology

Before diving into the analysis, the author defines NP as follows:

  • Naturalizing: Integrating phenomenological descriptions into the framework of natural sciences.
  • Phenomenology: Referring to both phenomenological philosophy and phenomenological psychology.

Neurophenomenology is based on the following working hypothesis:

Phenomenological descriptions of the structure of experience and corresponding descriptions in cognitive science should be related through Reciprocal Constraint (RC).

There are several theories within the NP research program, including:

  1. D. Dennett: Heterophenomenology
  2. E. Marbach: Formalization of Phenomenological Description
  3. S. Gallagher: Front-loading Phenomenology

The author evaluates Varela’s approach to Neurophenomenology as the most comprehensive, inclusive, and promising. Neurophenomenology emphasizes that phenomenological descriptions of experiences are essential for a coherent explanation of mind and consciousness. Furthermore, these phenomenological descriptions should be integrated into the explanatory framework.

Neurophenomenology posits that it is possible to create a complete and accurate description of first-person data, potentially in the form of mathematical models. While Husserl himself pointed out that the essence of conscious experience is not a precise morphological essence, the author argues, citing Park Seung-eok and Yoshimi, that this concern is not necessarily a barrier to developing mathematical methods. Thus, the success or failure of Neurophenomenology hinges on whether a mathematical model of conscious experience can be realized.

Three questions arise from the perspective of Neurophenomenology, though the author focuses only on the second and third:

  1. What exactly is the mathematical model that encompasses both the neuroscientific and phenomenological dimensions of consciousness? Can such a model exist at all?
  2. (Assuming the model exists) Can it provide a comprehensive bridge over the explanatory gap?
  3. (If it can bridge the gap) Is this approach compatible with Husserl’s stance? Is it not a form of naturalism, which Husserl rejected?

3. Bridging the Explanatory Gap

Varela’s core hypotheses can be classified into three major points:

1. Phenomenal consciousness cannot be reduced to physical events.
2. First-person descriptions and third-person research on conscious experience should be in a reciprocal relationship.
3. Existing first-person methodologies can be refined to adequately investigate phenomenal consciousness.

According to T. Bayne, there are different interpretations of the RC relationship mentioned in the second hypothesis:

 1. Reflective Equilibrium
 2. Heuristic Strategy
 3. Reciprocal Causation
 4. Generative Passage

Bayne finds the first two interpretations too weak and the third too strong, considering the fourth—the Generative Passage interpretation—the most appropriate. This interpretation views the RC relationship as a mathematical model connecting phenomenal states with neurophysiological states. Bridging this gap requires three elements:

 (1) Phenomenological descriptions of experiential structures.
 (2) A phenomenological dynamic model of these structural invariants.
 (3) The realization of these models within biological systems.

The problem can thus be redefined from the perspective of the Generative Passage interpretation:

 1. The Explanatory Gap as a scientific problem: The challenge of proposing a model that can coherently integrate phenomenology and neurobiology.
 2. The Hard Problem of consciousness in the philosophy of mind: The abstract metaphysical question regarding the status of consciousness within nature.
 (The possibility is acknowledged that one can solve the first problem without solving the second, or vice versa.)

Successfully bridging the explanatory gap, as redefined, involves (1) constructing a mathematical model that appropriately formalizes phenomenological events and neurological data, and (2) demonstrating that the phenomena to be explained (e.g., individual instances of consciousness or Neural Correlates of Consciousness, NCC) logically follow from the model.


4. Compatibility with Husserl’s Position

Naturalization can be broadly divided into (1) the naturalization of ideality and (2) the naturalization of consciousness. While Husserl opposes the former, he considers the latter possible within certain limits. The author argues that while the naturalization of phenomenological psychology as the foundation of empirical sciences is acceptable, transcendental phenomenology can never be naturalized.

"Explanatory Gap problem, Naturalizing Phenomenology, Neurophenomenology and Generative Passage interpretation." Generated by Midjourney AI

Reflections on the Reading

Naturalizing Phenomenology -> Varela’s Neurophenomenology -> Generative Passage Interpretation of Reciprocal Constraint -> Bridging the Explanatory Gap

This paper provided valuable insights into the integration of cognitive science and phenomenology. It clarified the NP research program, which is crucial in the philosophy of cognitive science. Moreover, it introduced Neurophenomenology as a promising theory within NP and discussed various interpretations of the core hypothesis—Reciprocal Constraint between first-person and third-person data. I came to understand that solving the explanatory gap between these data sources, as viewed through the Generative Passage interpretation, is key to constructing a significant mathematical model.

While studying Karl Friston’s Active Inference, a normative model of human behavior and psychological phenomena, I began to suspect that it could also effectively explain first-person data. Indeed, through interactions on Twitter, I discovered that Maxwell Ramstead, whose work I admire, has explored the philosophical stance on integrating Active Inference with phenomenology—Computational Phenomenology. I plan to read the following paper next to examine whether Active Inference could serve as the key mathematical model mentioned above:

Comments