But there is no strong argument for believing in this mind dependence. Galileo needed to only say that the color of a falling object is irrelevant to its place in physics, not that it has no color. In distinguishing between primary and secondary qualities Locke needed to only say that certain spatial configurations and movements of matter had the power to create sensations but not that the sensations have to exist "in us".
I'd argue that the dichotomy is false in the first place, even from a scientific point of view. Minds are ostensibly created by, and part of, nature. Sensations of color, depth, scent, etc. are all natural phenomena. The mental/physical dichotomy was created in the early modern period to help the natural sciences bracket off a whole off a whole set of philosophical questions so that progress could be made on the problems that were proving soluble. It's high time for that firewall to come down, but unfortunately it has become a calcified
dogma— something accepted uncritically and defended with religious zeal.
Corpusclarism, or approximations of it, also seem to be
naive, arising long before science and persisting in the face of contrary evidence from the sciences.
Empiricism gives pride of place to sensation. The models of our world that empiricism has produced has shown that at least the substance of these sensations is dictated by natural phenomena. That is, the
Hard Problem may remain, we do not have a good answer for "from whence sensation?" but the fact that changes in the environment can dictate changes in sensations, or that changes in the nervous system likewise affect sensation, even holding the environment constant, is as well established as anything. When you turn the light out, you can't see; likewise, when your eyes or visual cortex are damaged, you also cannot see.
Science has had to become like an eye trying to see itself. Notably for this analogy, that is something that is quite possible in the natural world; the eye can see itself using a reflection. Science is, in part, a system based on observations through which we try to untangle the how and whys of those same observations. Where things go awry is when
dogma declares that such sensations, the ones upon which all empiricism rests, are illusions, somehow unnatural. That is, as you say:
"objects are really colored." Yes, this is a
fact of the natural world.
What is interesting to me is that people seem to have an easy time dismissing scent, an ancillary sense for humans, as clearly illusory. "Shit smells bad to us and good to flies because scent isn't real." However, abstractions of space-time, e.g., volume, etc., based on a combination of visual, tactile, and vestibular sensations, are taken as iron clad facts of reality. It seems to me that, if we were dogs, we might go the other way on scent.
What is key though is that these senses help reinforce a unified model of reality. Senses appear to be used to cross check each other; when we can't tell if a flower is real or plastic by looking at it, we stroke and sniff it. This is why, while a mild degree of synesthesia seems adaptative, we still have extremely distinct sense impressions from different systems after billions of years of evolution (i.e., no one mistakes hearing for seeing)— one sense can cross check the veracity of another.
There is no
intrinsic reason to preference one sensory-based model or another. When we say "color is an illusion, the reality is photons moving through space," this is merely substituting an explanation of a reality in terms of an experience that only exists in one sense (color for sight) for a new model based on the sensation of movement through 3D space, something we experience primarily through three reinforcing senses.
Corpusclarism, allegedly dead in science, but seemingly very much alive, is the substitution of multiple models based on different types of sensory experience for a unitary model derived from the 'experience of / an analogy to', discrete objects in motion, something experienced in three senses. Thus: "Everything is objects in motion, sense experiences inessential to conceiving of objects in motion are illusory." Hence, "color is an illusion, volume and velocity are not."
I have to think that something unique to our biology (e.g., that our natural model of spacetime comes from multiple senses) is the motivation here because the entire concept of "discrete objects" has been challenged by physics, corpuscularism itself rebutted, and yet it remains the dominant way to envision the world. Not only that, but the concept is as old as written philosophy. It didn't take scientific revelations to make people buy into it, the position arose
ascientifically at the dawn of philosophy (e.g., a world of atoms, elements, Platonic solids, etc.), and has remained in vogue
despite countervailing scientific findings. That is, it seems to me that such views
are also naive forms of understanding.
What do we replace this seemingly inborne tendency with? That is a harder question. But it seems to me like a whole description of any given phenomena needs to include our sensation of it. Information, in the sense the term is used in physics, exists relationally. The current dogma involves an unnatural amputation of one half of the relation and the substitution of a magical, unnatural, "eye that sees things-in-themselves," in its place. Thus, neuroscience can inform our understanding of the illusory nature of things like color, but apparently not about the models that we have built atop those self-same illusory sensations?
A good start would be to try to knock down the tendency towards reduction that corpuscular thinking generates. A common view is that, not only is "color," not "real," but so to social status, recessions, etc. And yet a recession causes massive, observable physical changes across the globe, so in what way is it not "real" in the same sense that rocks are real?
Recessions
are real, but they are
incorporeal, a term from Augustine's day that is worth resurrecting. Incorporeal doesn't necessitate "non-natural" or "non-physical," although I will allow that the word is often used that way today. It means simply "lacking a body." (Corpus - Latin for body; "in" a Latin prefix for "un" or "not.")
Where is a recession located? In physical changes spanning the globe: in unfinished houses in Florida, in empty grocery shelves in Sudan, in government databases— trillions of digital logic gates— and in uncountable neurons and (natural) mental sensations. It exists substrate independently, in information that is naturally isomorphic vis-a-vis its encoding. That is, it is incorporeal, lacking a discrete body.
Color is likewise
incorporeal, it exists in patterns of neuronal activity, in the physical make up of stop signs, in the electrical currents powering lighting traffic lights, etc. It is impossible to study the science of marketing, traffic, the economics of entertainment, zoology, religious symbolism, etc., all natural phenomena without reference to color. Major problems in the social sciences could be resolved if it was acknowledged that, not only are the objects of study complex and emergent, but also incorporeal, but
incorporeal in a way that does not entail that they are "less real," or "non-natural."
The common sense view also says the Earth is flat and stationary.
A view that was overturned by careful
observations, i.e., by sensory experience. Note though that the new view, of a sphere in space rotating on its axis, moving through space around a larger sphere, is also a conception based in sensory experience. To be sure, more sophisticated models look at the system in purely mathematical terms, but mathematics itself is a discipline where
visualization is
more important than most. The axioms that ground proofs often come from appeals to (seemingly) essential truths about our visual or temporal-spatial sensations. That is, we accept Euclid's axioms because we cannot
visualize their violation coherently.
Sentience is a fuzzy term. But, to your point, slime molds demonstrate
intelligence when solving mazes, respond to stimuli, etc. without a nervous system. An interesting behavior of slime molds is that, when food runs scarce, numerous individuals will link together and form a migrating colony that can walk. Essentially, it is a composite body, a unity that appears to utilize a composite sensory system. As the name "
acellular slime molds" suggest, a subset don't even have proper cells.
Problem solving wise, they can do some neat things. People have used them to solve Hamiltonian path problems (e.g., the traveling salesman problem), a class of problem that gives digital computers a hard time. Although, DNA and RNA have also been used to "compute" answers to these problems by creating selection pressures that privilege shorter paths, but we don't think of those molecules as conceivably having any sort of sensory experience.
Seems to me like an assertion that will only get murkier as AI advances. Organisms lacking nervous systems nevertheless have sensory systems, so I don't see why the nervous system should be the deciding factor here.
If the core conceit of computational neuroscience is correct, then sensation is the result of sensory organs working in concert with computational processes. "Conscious" sensation would simply entail some given level of complexity and a "global workspace" in which sensory data is gathered and re-presented to another layer of
holistic computational processing. On this view, there is no reason why a sufficiently complex digital apparatus, or a primarily digital apparatus utilizing some biological components for parallel processing, could not experience conscious sensation.
But that makes sensation an information process, and such processes are generally considered to be substrate independent. So, if that conceit is correct, and it appears to be the number one theory, then you can theoretically build a mind that experiences sensation out of extremely well-crafted steam pipes that have been set up such that they mimic the informational structure of a human brain.
In any event, until there is a good theory for what
exactly causes sensation, this is an isolable debate (i.e., something more specific than "very complex information processing/computation processes," where "complexity," "information," and "computation," are all terms which generate hundreds of articles about how poorly defined and vague the currently are.)