Is it possible for a person with unlimited intelligence and no sensory experience to understand any concept like humans do? I am wondering if abstract information, information that has no reference with sensory experience, can be used to grasp concepts at a level similar to ordinary human understanding. What are the arguments for and against it coming from philosophers about it?
The particular quality of our sensory experience as human beings defines to a very large extent what it is to be a human being from a subjective perspective. We are, mostly, our perceptions. Include in these modes of perception, not only the five senses of vision, hearing, smell, touch and taste, but also proprioception, but also feelings, memory and imagination. The first five are modes of perception of the outside world. Proprioception is perception of the body. Feelings, memory and imagination are modes of perception of the brain itself.
Irrespective of whether the above is an accurate description, perception should be seen as modalities to acquire data about the world, our own body and our own brain being merely particular regions of the universe.
But, what is the use of those data? Clearly, our brain uses perception data to produce a sort of analogical model of our environment, model that includes not only a representation of the material world but also our own self as a sort of entirely fictitious homunculus subject.
However, this extraordinarily complex model doesn't serve any purpose. It just works. Human beings survive in this terrestrial world, and have developed an equally complex social structure now spanning the whole Earth.
What is human intelligence? Human intelligence is essentially a deductive model not of the world, but of reality. The model is best thought of as an analysis of the relations between perception data.
Our perception data, in evolutionary terms, were initially a sort of picture of our immediate environment. Our most basis perception senses, vision, hearing, smell, touch and taste, plus proprioception, provided initially and still provide today an immediate picture of our immediate environment. Memory and imagination provide a picture of an environment which is no longer so immediate. We remember our past, and places that may be far away because of the distance we may have travelled since we were there. We can imagine our future, possibly very far into the future, and places again far away from where we are.
Intelligence moves the boundary still further away from us here and now. Intelligence allows us to think not just our environment, not just the world, but reality itself, by building up not a picture, but an analysis of the relations between our perception data.
Thus, while our perception data define what it is to be human from a subjective point of view, it is our intelligence that defines what it is to be a human being objectively. It is the capacity that makes our kind of civilisation objectively possible.
Is it possible for a person with unlimited intelligence and no sensory experience to understand any concept like humans do?
Intelligence needs data about the world. No data, no intelligence. Our own sensory experience is essentially data about the world. What is special about our perception data is conceivably the quality of them, but whatever their quality they are still data. Pain is data about the health status of our own body, with built into it an emergency scale to signal to us how urgent it is to attend to a particular kind of damage.
Any non-human intelligence would have to have some analogous system of information about its environment and the state of its own material equipment. What would presumably be very different from us is the subjective quality of the data processed by this system. The world would look almost entirely different to it, qualitatively, than it does to us. However, it would use the data exactly in the same way as we do it, to build a picture of the world and an analysis of reality, only one dwarfing our own.
Perhaps one way to assess the difference with us is to look at the difference between us and other animals. Our species is very different from all other animal species. We certainly feel closer to species closer to us genetically, such as mammals generally, and primates, elephants and whales in particular. Yet, we can observe ants for example, and sort of understand their world, perhaps essentially because they are social animals, possibly gregarious if enjoyment is part of the range of feelings they can experience, if any at all.
But while we can to some considerable extent understand other species, it seems clear they can't understand us to anything like the same extent. An "unlimited intelligence" would probably be mostly impossible to understand for us.
A conversation would be possible, but only to the extent that this intelligence would choose to indulge our own very limited intelligence.
However, there is only one reality to be represented and analysed. Our own representation of the world would likely be very different from its own representation, but our own analysis of reality would likely be a simplified subset of its own. Our own concepts would be simplified analogues to its concepts. Its subjective experience, if any, would be very different, but the conceptual model of reality wouldn't be so utterly foreign to us.
We might even have interesting conversations about it.
This situation may not be so hypothetical as it sounds. AI machines, if we can make them really intelligent, and that is a very big "if" for now, would become a good approximation of an "unlimited intelligence". That is, not an actually unlimited intelligence but an intelligence so much greater than ours that it would look unlimited to us.
Maybe the real question is of the effect it would have on us to interact with an intelligence vastly superior to our own.