Posted by: rid November 2, 2013
Login in to Rate this Post:
0 
? 
?
Bostrom's Trilemma:
A Statistical Argument for Hindu Cosmology?
Some of the content of this article is taken from the book, Being and Perceiving
The Swedish philosopher Nick Bostrom has argued that, statistically, we are more than likely living inside a computer simulation. In Hinduism, the phenomenal world is said to be a mere dream in the mind of Brahman. I suggest that Bostrom's "simulation hypothesis" may provide a statistical argument for Hindu cosmology.
Essentially, Bostrom argues that if humans ever reach a state of such technological sophistication that we can run computer simulations of our past, then we will run many such historical simulations. In such a scenario, there would be only one genuine physical reality, but perhaps billions of simulated realities. It is therefore statistically probable that we are currently inhabiting such a simulation.
Bostrom phrases his argument in terms of a trilemma. In his own words:
"at least one of the following propositions is true: (1) the human species is very likely to go extinct before reaching a 'posthuman' stage; (2) any posthuman civilization is extremely unlikely to run a significant number of simulations of their evolutionary history (or variations thereof); (3) we are almost certainly living in a computer simulation."
Sensory Awareness and the Phenomenological World
In Hindu cosmology, the phenomenal world is believed to be a dream in the mind of Brahman - the supreme deity and ground of all being. Accordingly, the aim of Hindu practice is to free the devotee from the illusion of the phenomenological realm by uniting their consciousness with that of Brahman.
The phenomenological world is made up of qualia, which are units of sensation (greenness, hardness, loudness, etc.). Phenomena are conglomerations of qualia (e.g., an apple may be a combination of roundness, greenness, firmness, smoothness, etc.). With the exception of man-made objects, objects do not exist independently of their qualities.
It should be noted that I am not ascribing objective existence to qualia - softness does not exist independently of soft objects, for example. Qualia are the phenomenological units from which our sensations are composed, and as such they are entirely psychological.
The process of decoding neural representations consists of deriving qualia from sensory input, and is thus the process by which the phenomenological world is constructed from brain activity (for more on this, see my article The Hard Problem of Consciousness Explained, and my book, Being and Perceiving).
Simulated Awareness
Imagine a computer program which can scan visual input (e.g., from a camera) and identify faces in that input. In order to perform this function, it must possess a template of a face which the input can be matched against; this is the archetype of a face. The computer thus has both a limited perceptivity (the ability to scan its visual field) and the ability to see faces in the data. Together, these two things constitute a limited level of sensory awareness; the computer must be able to literally see faces.
If the ability to scan were increased to take in the whole visual field at once, and the computer was also endowed with the ability to identify geometry and perspective (and thus objects and their position in space), motion (direction, speed, etc.), light and dark, and so on, it would be endowed with the ability to derive qualia from visual input, and would thus possess awareness of a phenomenological space. It would be able to interpret visual input as a phenomenological world, and would thus be able to see. In contrast, to be unable to distinguish between shapes, sizes, distances, etc., would be blindness.
Once sensory awareness is explained in these mechanistic terms, then all the other functions of consciousness - abstraction, reflection, self awareness, etc. - can be explained through a recourse to this initial layer of sensory processing.
If a computer program can theoretically simulate sensory awareness, then one computer can also simulate two or more interacting consciousnesses. Effectively, numerous simulated minds could exist inside a computer, interacting with each other and with a simulated phenomenal world. They would have no idea that their perceptions were not derived from an underlying physical reality, but from numerical code.
Sentience Quotient
The psychologist and physicist Robert Freitas defined sentience as the information processing capacity of a given system. He also developed the concept of the sentience quotient (SQ - analogous to IQ), which assigns a value to the sentience of an information processing system, based on the relationship between the size of each processing unit, processing speed per unit and the overall number of units.
Humans and other animals, computers, and even plants and fungi, can all be assigned a sentience quotient, since all process input and generate output. A venus flytrap has an SQ of +1, while the theoretical upper limit to the scale is +50. Humans have an SQ of +13. Other mammals are only handful of SQ points away from human beings.
This begs the question, why should a sufficiently sentient being not also be able to consciously simulate phenomenal realities, simply by imagining a virtual reality to a degree of detail comparable to a computer simulation? For example, imagining the interactions of individual particles and the firing of neurons in a multitude of brains? Would the imagined beings not experience themselves as possessing a genuine existence?
This may seem fantastical, but we should remember that the gap between human sentience and the highest possible sentience is nearly three times greater than the gap between a human being and a plant.
Consider: the brain is essentially a computer. Computer simulations run on electrical signals, just as the brain does. A human brain is capable of generating an entire phenomenal world. Why should a brain many times more powerful not be capable of generating numerous interacting minds, set within a phenomenal universe?
This puts another spin on Bostrom's trilemma. If a race of beings this sentient were to evolve, statistically we are most likely to be dreams in their consciousness. In effect, this is a statistical argument for Hindu cosmology.
Let's refer to such a race of beings, possessed of this level of sentience, as a Brahmanic species. The evolution of such a species may seem unlikely compared to the computer simulation variant of the trilemma. However, posthumanism may well render the distinction negligible. Perhaps we genuinely are just dreams in the mind of Brahman.
Copyright (c) Daniel E. Haycock 2011, all rights reserved.
A Statistical Argument for Hindu Cosmology?
Some of the content of this article is taken from the book, Being and Perceiving
The Swedish philosopher Nick Bostrom has argued that, statistically, we are more than likely living inside a computer simulation. In Hinduism, the phenomenal world is said to be a mere dream in the mind of Brahman. I suggest that Bostrom's "simulation hypothesis" may provide a statistical argument for Hindu cosmology.
Essentially, Bostrom argues that if humans ever reach a state of such technological sophistication that we can run computer simulations of our past, then we will run many such historical simulations. In such a scenario, there would be only one genuine physical reality, but perhaps billions of simulated realities. It is therefore statistically probable that we are currently inhabiting such a simulation.
Bostrom phrases his argument in terms of a trilemma. In his own words:
"at least one of the following propositions is true: (1) the human species is very likely to go extinct before reaching a 'posthuman' stage; (2) any posthuman civilization is extremely unlikely to run a significant number of simulations of their evolutionary history (or variations thereof); (3) we are almost certainly living in a computer simulation."
Sensory Awareness and the Phenomenological World
In Hindu cosmology, the phenomenal world is believed to be a dream in the mind of Brahman - the supreme deity and ground of all being. Accordingly, the aim of Hindu practice is to free the devotee from the illusion of the phenomenological realm by uniting their consciousness with that of Brahman.
The phenomenological world is made up of qualia, which are units of sensation (greenness, hardness, loudness, etc.). Phenomena are conglomerations of qualia (e.g., an apple may be a combination of roundness, greenness, firmness, smoothness, etc.). With the exception of man-made objects, objects do not exist independently of their qualities.
It should be noted that I am not ascribing objective existence to qualia - softness does not exist independently of soft objects, for example. Qualia are the phenomenological units from which our sensations are composed, and as such they are entirely psychological.
The process of decoding neural representations consists of deriving qualia from sensory input, and is thus the process by which the phenomenological world is constructed from brain activity (for more on this, see my article The Hard Problem of Consciousness Explained, and my book, Being and Perceiving).
Simulated Awareness
Imagine a computer program which can scan visual input (e.g., from a camera) and identify faces in that input. In order to perform this function, it must possess a template of a face which the input can be matched against; this is the archetype of a face. The computer thus has both a limited perceptivity (the ability to scan its visual field) and the ability to see faces in the data. Together, these two things constitute a limited level of sensory awareness; the computer must be able to literally see faces.
If the ability to scan were increased to take in the whole visual field at once, and the computer was also endowed with the ability to identify geometry and perspective (and thus objects and their position in space), motion (direction, speed, etc.), light and dark, and so on, it would be endowed with the ability to derive qualia from visual input, and would thus possess awareness of a phenomenological space. It would be able to interpret visual input as a phenomenological world, and would thus be able to see. In contrast, to be unable to distinguish between shapes, sizes, distances, etc., would be blindness.
Once sensory awareness is explained in these mechanistic terms, then all the other functions of consciousness - abstraction, reflection, self awareness, etc. - can be explained through a recourse to this initial layer of sensory processing.
If a computer program can theoretically simulate sensory awareness, then one computer can also simulate two or more interacting consciousnesses. Effectively, numerous simulated minds could exist inside a computer, interacting with each other and with a simulated phenomenal world. They would have no idea that their perceptions were not derived from an underlying physical reality, but from numerical code.
Sentience Quotient
The psychologist and physicist Robert Freitas defined sentience as the information processing capacity of a given system. He also developed the concept of the sentience quotient (SQ - analogous to IQ), which assigns a value to the sentience of an information processing system, based on the relationship between the size of each processing unit, processing speed per unit and the overall number of units.
Humans and other animals, computers, and even plants and fungi, can all be assigned a sentience quotient, since all process input and generate output. A venus flytrap has an SQ of +1, while the theoretical upper limit to the scale is +50. Humans have an SQ of +13. Other mammals are only handful of SQ points away from human beings.
This begs the question, why should a sufficiently sentient being not also be able to consciously simulate phenomenal realities, simply by imagining a virtual reality to a degree of detail comparable to a computer simulation? For example, imagining the interactions of individual particles and the firing of neurons in a multitude of brains? Would the imagined beings not experience themselves as possessing a genuine existence?
This may seem fantastical, but we should remember that the gap between human sentience and the highest possible sentience is nearly three times greater than the gap between a human being and a plant.
Consider: the brain is essentially a computer. Computer simulations run on electrical signals, just as the brain does. A human brain is capable of generating an entire phenomenal world. Why should a brain many times more powerful not be capable of generating numerous interacting minds, set within a phenomenal universe?
This puts another spin on Bostrom's trilemma. If a race of beings this sentient were to evolve, statistically we are most likely to be dreams in their consciousness. In effect, this is a statistical argument for Hindu cosmology.
Let's refer to such a race of beings, possessed of this level of sentience, as a Brahmanic species. The evolution of such a species may seem unlikely compared to the computer simulation variant of the trilemma. However, posthumanism may well render the distinction negligible. Perhaps we genuinely are just dreams in the mind of Brahman.
Copyright (c) Daniel E. Haycock 2011, all rights reserved.