Introduction to Evolutionary Epistemology as a Philosophical Position
| No Previous Article | Back to Evolutionary Epistemology Index | Spirituality in Evolutionary Epistemology |
Traditional Epistemology
Epistemology is the branch of philosophy which studies the nature of knowledge and its acquisition.
Traditional epistemology is anthrocentric. Its focus is human knowledge and, in particular, knowledge which can be expressed in language.
Following the example of Plato, the most common expression of traditional epistemology is the identification of knowledge with true, justified belief. After all, it is argued, that if Pat knows that snow is white:
- Pat must believe that snow is white
- it must be true that snow is white and
- Pat must be justified in believing that snow is white.
This identification spawns a new set of questions:
- What is belief?
- What is truth?
- What is justification?
In addition, there is a central theme of regarding belief as belief in the truth of a sentence. But what if Pat only speaks German? Pat may then believe that "Schnee ist weiss", but does she believe that "Snow is white"? The traditional solution is to postulate propositions as the meanings of sentences. "Schnee ist weiss" and "Snow is white"? are said to express the same proposition. If so, we can say the Pat believes (the proposition) that snow is white even though she doesn't speak English.
There is, of course, some skepticism about the existence of propositions. This gives rise to additional questions about the nature of meaning.
These questions dominate traditional epistemology and the various answers have met with a host of difficulties.
Before we leave traditional epistemology, it is worth noting that some traditional epistemologists have recognized the existence of non-propositional forms of knowledge.
There is, for example, the difference between knowing how to ride a bicycle and knowing that a bicycle is capable of being ridden. That is, there is a difference between knowing how and knowing that, and only the latter could be considered propositional knowledge.
Some philosophers, Bertrand Russell for example, distinquish between knowing yellow and knowing that the sun is yellow. Russell calls the first "knowledge by acquaintence" and the second "knowledge by description", where again, only the latter could be considered propositional knowledge.
When we learn to use the word "yellow", we presumably do so by having repeated experiences of yellow together with some authority repeating the word "yellow" on those ocassions. If so, knowledge by acquaintance is more fundamental than knowledge by description, that is, propositional knowledge.
Evolutionary Epistemology
In contrast to traditional epistemology, evolutionary epistemology regards knowledge as something that evolves as life itself evolves. Evolutionary Epistemologists see a continuity between animal and human knowledge and between child and adult knowledge. They require that an adequate epistemology explain how one arises out of the other.
Evolutionary Epistemologists do not see knowledge as fundamentally linguistic. Linguistic knowledge arises out of non-linguistic knowledge. An adequate theory of knowledge needs to be able to support both linguistic and non-linguistic knowledge. In Evolutionary Epistemology, knowledge is a representation or a map, in exactly the sense that a roadmap is a representation of an external reality of roads.
By regarding knowledge as representational rather than propositional, the question of truth becomes less binary. A proposition is either true or false. A representation is more or less accurate, more or less precise. We will sometimes be concerned with whether knowledge is adequately accurate or adequately precise rather than whether knowledge is true (100% accurate and 100% precise).
These concepts are also more intelligible when applied to knowing how and knowledge by acquaintance. One can more or less know how to ride a bicycle. One can more or less know how distinguish shades of yellow from shades of orange or to distinguish one shade from another. It makes sense to say that a champion cyclist has more knowledge of how to ride a bicycle. It makes sense to say that a visual artist has more knowledge of yellow than a non-artist.
These concepts are capable of making sense of a claim that one can know that Neverland is mountainous, that is has a deep water harbor and has islands off the mainlland. But the concepts are problematic when making sense of a claim that round squares are impossible.
The Basis of Evolutionary Epistemology
It might seem obvious that the scientific basis of Evolutionary Epistemology is biological. However, physicists, chemists and sociologists also have a claim. We consider their claims and the claim of biologists in what follows.
Physics as the Basis of Evolutionary Epistemology
Historically, the claim of physics to be the basis of Evolutionary Epistemology can be traced back to a book by the physicist Erwin Schrödinger published in 1944 entitled "What is Life?" [SE1.1944a].
Schrödinger is perhaps best known as one of the fathers of Quantum Mechanics. In fact, the fundamental equation of Wave Mechanics (which together with Matrix Mechanics is one of the two main versions of Quantum Mechanics) is called "the Schrödinger equation", after Schrödinger.
In "What is Life?" Schrödinger notes that the physical laws governing large scale phenomena are the result of chaos on a smaller scale. This is the principle of "order from disorder." One example he produces is diffusion, which appears as a highly ordered process caused by an underlying random movement of molecules. As the number of particles in a system is reduced, the behaviour of the system becomes more random.
He believes that life depends on the existence of order. Given the complexity of life, it might be naïvely assumed that the master code of a living organism must contain a large number of particles. In reviewing what was then known about the mechanism of heredity, he asserts, contrary to the naïve assumption, that what carries hereditary information must be small in size and stable over time.
Schrödinger's candidate for the small, stable carrier of hereditary information is a molecule, which can be stable even if consisting of only a few atoms. Although the existence of molecules were long known, their apparent stability is inexplicable in classical mechanincs. In classical mechanics, even atoms were unstable. Quantum mechanics, on the other hand, convincingly explains the stabillity of both atoms and molecules. Schrödinger also assumed (incorrectly) that Darwin's needed mutations are due to a quantum effect.
Schrödinger believed that the heredity material is a molecule whose structure does not repeat itself. He calls this type of molecule an "aperiodic crystal". Because it is aperiodic, it can encode large number of possibilities with only a small number of atoms.
---STUB:Physics completed
Chemistry as the Basis of Evolutionary Epistemology
---STUB
Biology as the Basis of Evolutionary Epistemology
---STUB
Sociology as the Basis of Evolutionary Epistemology
---STUB
The Evolution of Knowledge and Meaning
Here we give a sketch of the position we wish to defend concerning the evolution of knowledge and meaning. This sketch consists of the series of steps which we believe were required for knowledge and meaning to evolve.
Before Knowledge and Meaning
There are a number of good reads which tell the story up to the point at which we wish to begin. Going back to to the absolute beginning is "At the Heart of the Web"[SR1.1989a] subtitled "The Inevitable Genesis of Intelligent Life", by the physicist, cosmologist and astronomer George A.. Seielstad. A more detailed account of how life could arise from non-life from the perspective of statisitical thermodynamics is physicist Ronald F. Fox's "Energy and the Evolution of Life" [FR1.1988a]. For a biologist's perspective, see "Four Billion Years: An Essay on the Evolution of Genes and Organisms" [LW1.1988a] by William F. Loomis. Finally, there is a very accessible work for the lay person by biologist Richard Dawkins called "The Ancestor's Tale" [DR1.2005a], and subtitled "A Pilgrammage to the Dawn of Life".
The Origins of Knowledge and Meaning
We begin our account at the point in earth history when there are multicellular organisms whose cells are differentiated and and the different types of cells perform different functions for the organism as a whole. Prior to this point, knowledge and meaning do not exist. At some point an organism evolves which has a simple sensory anatomy. The diagram below shows such an organism schematically.
As a knower, this organism has three basic components:
- a sensory receptor, capable of responding to an input from the external world and passing its response on to a sensory processor,
- a sensory processor, capable of transforming an input from a sensory receptor and passing that transformed input to a sensory representation,
- a sensory representation, capable of encoding information which represents some feature of the external world.
We make the following claims about such an organism:
- its sensory representation is knowledge of the external world,
- the meaning of its sensory representation is the feature(s) of the external world it represents,
- this knowledge and meaning are the most primitive forms of knowledge and meaning respectively.
These claims may seem counterintuitive for the following reasons:
(Reason 1) Claim 1 is not knowledge in the sense of true, justified belief. That is, there does not seem to be any belief involved since the organism is too simple to be conscious of its representation. There is no justification involved; the organism is simply caused to have its representation. The representation may not even be true since it could result from a trauma or the interventation of an experimenting scientist.
(Reason 2) Claim 2 is not meaning in the sense of reference; the organism's representation does not mean anything like what we mean when we say the sentence "Snow is orange" means that snow is orange.
We agree that Claim 1 is not about knowledge in the sense of true, justified belief. We are proposing a use of the word "knowledge" which is non-traditional. We do so because we seek a concept of knowledge which allows for a continuity between animal and human knowledge and between primitive and sophisticated knowledge. Could we use another term like "proto-knowledge" or "proto-belief" for what we are calling knowledge? Of course we could. We would if we needed to refer to knowledge in the traditional sense. But given that we cannot know that we know anything in this sense, we do not have any immediate use for this concept. Should we have occasion to refer to knowledge in the traditional sense, we will use the term "true, justified, linguistic knowledge".
We also agree that Claim 2 is not about reference in the traditional sense. In the traditional sense, we learn that "dog", for example refers to a number of creatures through a short training procedure. Yes, Fido is a dog. No, Fluffy is not a dog; Fluffy is a cat. No cats are not girl dogs.
Our use of "reference" is more fundamental than the traditonal use in the sense that the traditional use depends on the prior existence of reference in our sense. Our use is of reference as a mapping between the representation (which is itself a map) and the feature mapped. This mapping is a one-to-one correspondence. This mapping is established by the evolution of organisms with a anatomy which generates the representation when its sensory receptors are presented with the represented feature under normal conditions. That is, the mapping exists because, under normal conditions, the presence of the represented feature causes the representation.
The existence of the representation does not guarantee the existence of the represented feature. It might have been caused by a blow or interference by an experimenter. It is not the existence of the represented feature that interests us here; it is the existence of the mapping, which is established by the structure of the organism (it's anatomy) and the causal structure of the world. Further, the existence of the mapping establishes the meaning of the representation as a reference to the represented feature.
Encoding Representations
Every map uses an encoding schema.
A roadmap uses lines of various types to represent roads of various types. It uses blue lines to represent rivers. It uses paired brackets, ][, to represent bridges. Closed and open circles of various sizes represent villages, towns and cities.
A perceptual representation uses an encoding schema of electrical potentials. The charge in a car battery is one example of an electrical potential.
In an example of simple visual representation, the electrical potential in a cell represents the presence of a light source, the higher the potential, the brighter the source. The cell electrical potential is not a perfect representation of the brightness of the light source The cell has a maximum potential and a number of discrete levels which do not perfectly match the discrete levels of brightness of the light source. Suppose the cell has ten discrete levels which map to ranges in the brightness of the light source as follows:
| Brightness | Cell Potential | |
|---|---|---|
| Minimum | Maximum | |
| 100.00 | ∞ | 9 |
| 99.50 | 99.99 | 8 |
| 89.00 | 94.99 | 7 |
| 78.25 | 89.99 | 6 |
| 67.50 | 79.99 | 5 |
| 56.70 | 74.99 | 4 |
| 46.00 | 66.99 | 3 |
| 35.00 | 59.99 | 2 |
| 24.00 | 49.99 | 1 |
| 0 | 23.99 | 0 |
So each level has a minimum brightness which will trigger it. Below 24.00, it is not triggered at all (that is, it is 0). Everything above 100.00 triggers the maximum potential. The representation can now be expressed as a number, each digit corresponding to a range of brightness.
Suppose now that our organism, instead of having a single visual receptor which responds to all wavelengths of light, has three receptors which each respond to a different range of wavelengths. In organisms with trichromatic vision (like humans), these ranges are wavelengths of reds, blues and greens. Because there are three color receptors, there will also be three representation cells, one corresponding to each receptor.
The representation of our single, all wavelength receptor organism can be described by a single number. Describing the representation of a three color receptor organism requires three numbers in some specified order, that is, an order set of three numbers. The order is important. If the order is red first, then green, then blue, <9,0,0> represents red, <0,9,0> represents blue, and <0,0,9> represents green. <9,9,9> represents white and <0, 0, 0> represents black.
In mathematics, an order set of numbers is called a "vector". Given the allowed values for each position in the ordered set, we can define a "vector space" which is a container for every allowed vector. Since there are 10 possible choices for each position, the number of possible vectors is 10 X 10 X 10 or 1,000. Each is the representation for a different color. The diagram below shows the vector for yellow, <9, 9, 0> in this space.
For the sake of simplicity, we show the dimensions as having 10 discrete units. In humans, the number is much higher. A computer typically uses 256 units per dimension, giving a total of 16,777,216 colors
This example is also an example of trichromatic representations. Some animals, bees for example, have tetrachromatic representations, that is they have four types of color receptors. (Bees can see infrared in addition to red, green and blue). The vector space for tetrachromatic vision is necessarily four dimensional. In humans, there are five dimensions in the space for taste receptors: sweet, sour, bitter, salty and savory (or umami).
Multimodal Knowers
Of course, organisms respond to more than just light. The following table gives modalities currently studied in humans. The first five are the five traditional senses.
| MODAL SOURCE | COMMON NAME | SCIENTIFIC NAME |
|---|---|---|
| pressure | touch | somatosensation |
| light | sight | vision |
| sound | hearing | audition |
| chemical vapor | smell | olfaction |
| chemical compounds | taste | gustation |
| temperature | thermoception | |
|
kinesthetic sense (perception of positions of body parts) |
proprioception | |
| pain | nociception | |
| balance | equilibrioception | |
| vibration | mechanreception | |
| others | see Wikipedia article on Sense |
In addition to human modalities, other animals can perceive electrical or magnetic fields or experience sound more like our experience of sight (echolocation in bats and sonar in cetaceans).
The following diagram shows the anatomy of a two sensory mode processor.
Sensorimotor Knowledge and Meaning
The next organism we wish to consider is a sensorimotor knower.
---STUB: completed document
Document Control:
|
Author: Gilbert Bruce Fargen |
|
|
Date of First Publication: 22 Jan 2017 |
This website has adopted and enforces a Privacy Policy. |
|
Date of Last Revision: 22 Jan 2017 |
Use of material on this page and this website in general is subject to the Web Use Agreement. |
