David Baird “The Unexpressible Knowledge Argument”

I will discuss versions of the knowledge argument. The official version is that we raise a scientist, Mary, in a black-and-white room, and teach her “everything about the physical properties of physical objects.” And then on her lucky day, we release her into the world of color, and she learns what it is like to see red. So there must be non-physical properties, and physicalism is false.

However, this story is implausible, because even in the room, Mary would be able to see red, simply by closing her eyes, due to the electrochemical fluctuations in the interaction between her retinas and her visual cortex. So we have to change the story. We could give her an anti-color drug that would keep her in the black-and-white world, but this conflicts with one of the aspects of the discussion, that maybe Mary could invent red, given the proper information. The drug would interfere with this.

My improved version is to raise a scientist, Noel, in the normal world, and eventually give her LSD, a psychadelic drug, after she has been thoroughly educated about “the complete neuroscience of the human brain, off and on LSD.” Noel then learns what it is like “to trip”, which is what the LSD-influenced experience is called. Unlike seeing red, tripping is not part of the

normal functioning of human consciousness, and it isn’t something that could just spontaneously emerge from normal neural processes.

I put quotes around “the complete neuroscience of the human brain, off and on LSD”, because as I will try to show, a real “science” of the brain and its “properties” involves more than information that is expressible. The same goes for Mary. To know all the physical properties of physical objects such as brains when seeing red would involve knowing information that is not expressible, unless adequate triggers like red objects are used. (A trigger is a thing that is observable—a thing that “triggers” our minds into going through perceptions).

What makes one begin to doubt the knowledge argument’s strength is that even if physicalism is true, Mary and Noel both learn something new when given their triggers. This is the case, because no matter what they learn about the physical, the key knowledge is unavailable until the key experience is given to them. At least with Noel. I should stop talking about Mary, because since color is such an intrinsic and normal part of human perception, she could possibly invent red (imagine it internally) on her own, through her cleverness, or the power of the black-and-white education she receives. This eventuality would invalidate the knowledge argument, which I want to do, but not like this. This just highlights the fact that the Mary version is unrealistic. With Noel, we can be positive that her invention of tripping impossible, due to its intensity, which is highly extreme, although unpredictably good, bad, or mixed.

There’s only one way to learn what it’s like to trip. Noel has to trip. She can have objective knowledge of a tripping brain, but she needs subjective knowledge. It’s necessary. But it can’t be expressed to her or shared with her. It’s selfish, or selfist. She can’t see it by studying a brain model. We give Noel video of people tripping, poetry written by them, papers on the philosophy of LSD, investigations into the neuroscience of tripping, a computer simulation of a brain on LSD. People who are tripping are introduced to her and she can touch them and look in their eyes. Yet all of this misses tripping. She still doesn’t know what it’s like, exactly. Of course, when she trips, she may say, “Oh, now I see what you were talking about.”

The experiment proves Noel learns a subjective fact, not necessarily a non-physical fact. To call the subjective non-physical is begging the question. While a subjective state may not be expressible to other people, may not be compatible with representation by an external physical object, this subjective state is still observable, if only by one subject (mind). Indeed, the subjective is closer than the objective to the only reality we can know, our consciousness. What sense does it

make to judge that only things others can see are real, are physical? If you really only think true knowledge that is objective, then you’d have no way to think about other minds, because no matter how closely you look at peoples’ brains, you can’t ever see their subjectivity. You’d have to see minds as only objects. When really it’s the other way around—objects are only subjectively knowable. This doesn’t mean we can’t learn about them with other people. This intersubjectivity is an amazing way to learn.

We tell Noel “The brain does x when tripping”. But this doesn’t mean her brain can do x without LSD. We tell Noel “A trip is something you haven’t had any experience with. It will be different than anything you’ve ever done”. This is vague, yet accurate. How should we describe a trip?

What the brain does during it? But “what the brain does” is both objective and subjective. We have to know the subjective to have a full picture. We can express with respect to subjectivity, but can we express subjectivity? An objective thing can’t know subjectivity unless it becomes subjective. Can we use objective info to create subjective states? All expression does is to objectify subjectivity. Some truth can be objectified, some can’t.

Indeed, here is no objective/physical color red—it is a quality that only exists in consciousness. Photons of wavelength x are not “red”—our brain is programmed to make them “look” red. An alternative color could have been assigned to these wavelengths. Indeed, we could hear ‘light’, and ‘see’ sound. If red can’t be expressed non-visually, then of course we can’t tell Mary what it’s like to see it. But what if it could be expressed, say, verbally? Can we feed her subjective information? Can we hook her brain up to a dream-causing device?

What about a drug no one has used? Could we predict its effects? Using an ultra-detailed brain simulator? The simulator would have to be conscious. Giving it the drug would need to be no different than giving a person the drug. So for all the truth to be known, someone has to use the drug. So certain unique physical triggers are necessary for knowledge. There is knowledge only a subject can know. There is knowledge such that no amount of objective science can predict how it is to know it.

The tripper learns not only the trip, but a new paradigm, a point-of-view, from which all knowledge is different. Objective facts are all changed by a changed perspective on them. A perspective is the integral of all info, but is it info? There are multiple ways/methods of integrating conscious material. The tripper is given no new info. So objective info is not always necessary for learning. We learn from our subjectivity.

The pre-tripped Noel has access to any info, no matter how colorful, indexical, subjective, right/wrong, sensual, factual. So we can automatically say facts aren’t sufficient for certain perspectives. Of course Mary can’t see red. We haven’t set up an environment sufficient for seeing red. We’ve given her the wrong info. We’ve lied to her. “This is all there is to know.” But this is not all there is. Words are things, like red objects are things. Things are objective. Physical knowledge is of things. The color red is subjective, but its trigger is a “red” object. Red is a property of consciousness. Mary is learning about consciousness. She’s learning some thing about her self. She’s been given world-facts, but not all the self-facts.

To learn the subjectivity of red, Mary has to go through something. The sufficient thing. Something to change her brain. We can let Noel go through anything she wants, except taking LSD, “dosing”. Only dosing can “tell us about dosing”. Only red can “tell us about red.” We have to learn both subjective and objective facts. And all knowledge is from triggers.

Every new “what it’s like to feel x” is a new fact. Every new moment is a new subjective fact. Noel can learn without tripping. So why is it surprising that she learns with tripping? What does learning something tell us? We could teach Noel “all” the subjective facts with a brain-print device. So then what would her next learned fact be called, according to a knowledge argument? Nonmental? Metametaphysical? Subjective-subjective?

What’s more interesting, some things need triggers again, even if already “known.” Some facts can only be known temporarily. Mary could see red, then forget what it’s like. Hard for us to do, since we’ve seen it ten million times. Vividness is no guarantee facts will/can be remembered. If she can’t remember red, could adequate “code” be fed into Mary’s “program” for an invention of redness? How flexible are our programs? What is learnable? Could Einstein invent red?

I think the key to this discussion is the question What is “what it’s like to x”? Just what a physical brain is programmed to feel? Why not say the mental is physical? We’ve shown that to know everything, Noel and Mary need to know the subjective as well as the objective, the mental as well as the physical. And for some learning, specific triggers are needed, like red things and LSD. The knowledge argument gets it wrong because its first premise, that Mary knows “about” red, ignores that fact that she needs to know red to know all the properties of objects. And also, one object we want to learn about is the mind, which has both subjective and objective properties, both of which have to be learned to have complete knowledge. Because something can only be known subjectively, that doesn’t mean it’s nonphysical. Indeed, the subjective point-of-view is

how we known everything, including physics. To call the mental nonphysical is then equivalent to calling everything, including the physical, nonphysical.