In his paper Epiphenomenal Qualia Frank Jackson invites us to consider the imaginary case of Mary, kept in a monochromatic room from birth and who, presumably out of boredom, spends her time becoming acquainted with all that neuroscience can tell us regarding the mechanisms that underlie our experience of colour vision. Mary herself has never seen a red object, but when it comes to the physical facts that attend such an experience, she knows them all. What, Jackson asks, would happen were she to be released from her room and to see a red object for the first time? Would she learn something new? Surely she would: she would learn what the experience of seeing a red object is like. But in that case would it not follow that, since she already knew all the physical facts about “seeing red”, what she learns must be a “non-physical fact” (a fact not present in the developed neuroscience of colour vision)? And if there are such “non-physical” facts does it not follow that physicalism is false?
Jackson’s “knowledge argument” against physicalism has its detractors. These days they include Jackson himself who has, in his own phrase, “capitulated” to the orthodoxies of scientific materialism. More generally we can ask the question: what should we do when the results of a thought experiment are inconsistent with the prevailing view? Abandon the thought experiment? Or abandon the prevailing view? And is there a core philosophical principle, some piece of metaphilosophy, to which we can appeal in order to settle the matter?
Some argue that thought experiments tend to confuse what is possible (in the conceptual sense) with what is imaginable (in the epistemic sense). Thus Hilary Putnam: you can imagine that you can imagine that you are a “brain in a vat” and that what you take to be your thoughts and sensations are a collective and systematic misrepresentation of the world “as it really is” but what you imagine you imagine is in fact no such thing. A brain in a vat would not be able to imagine itself to be a brain in a vat: the very notion is incoherent (an implication, Putnam argues, of a true theory of meaning). Similarly Bernard Williams has suggested that thought experiments which invite us to imagine ourselves in this or that scenario will often overlook that we are embodied originators of our own projects and goals, and that this fact describes an ineliminable feature of our personal identity. Thus when John Rawls constructs his version of political liberalism via the imagined consent of a freely choosing rational agent operating from behind a “veil of ignorance”, Williams will point out that when the free agent takes the veil she ceases to be an agent at all.
Small wonder then that the thought experiment passes in and out of fashion. The late philosopher of mind and freedom activist Kathy Wilkes described her own book on personal identity as being philosophy without thought experiments. John Searle in a series of exchanges with Paul and Patricia Churchland complained that the problem with thought experiments was their failure to preserve the philsophically salient features of the problem they are intended to illuminate. Pretty cheeky perhaps, from the inventor of the Chinese Room. But not without chutzpah.
Some of the great exponents of the thought experiment appreciate also the value of the genuine experiment. None more so than Berkeley who on one occasion attempted to hang himself in order to generate a near death experience. The episode is described by Oliver Goldsmith who noted that it was agreed that:
….his companion would take him down at a signal agreed upon…Berkeley was therefore tied up to the ceiling, and the chair taken from under his feet, but soon losing the use of his senses, his companion it seems waited a little too long for the signal agreed upon, and Berkeley had like to have been hanged in good earnest; for as soon as he was taken down he fell senseless and motionless upon the floor….
So take note: the next time you hear of a pop star or actor who has passed away in embarrassing circumstances things might not be so tawdry as first appears. Their final thoughts might have been deep ones.
In mathematics one sometimes is able to make use of exaggerated cases to get a better handle on a conjecture. The solution to the exaggerated case is not the answer to the conjecture and isn’t meant to be. It’s used to open some doors that are otherwise bolted.
I like thought experiments, and they work on me the same way as the above described situation in mathematics. True, in the several t.e.’s mentioned in the post, I was left with questions, questions I feel motivated to find related information on. If one uses the t.e. in a comparable manner I believe it can be a valuable tool as well as being fun.
Andy Walsh asks:-
“what should we do when the results of a thought experiment are inconsistent with the prevailing view? Abandon the thought experiment? Or abandon the prevailing view? “ and “is there a core philosophical principle, some piece of metaphilosophy, to which we can appeal in order to settle the matter?”
I would say you should do what any scientist would do when faced with scientific results inconsistent with the prevailing view. Abandoning the experiment or the prevailing view out of hand is surely bad scientific/philosophical method. All science is tentative and surely all thought experiments are similar. In the light of further knowledge we constantly alter and refine what we already believe. It is not only thought experiments which eventually turn out defective in some way.
Frank Jackson’s “What Mary Didn’t Know” was published in(1986). At that time she acquired, we are led to understand, all the physical information there is to obtain about what goes on when we see ripe tomatoes. My point is here how could it have been known at that time, or any other time for that matter she had acquired ALL the possible knowledge. Certainly much more concerning the visual processes has been discovered since.
Some science and some thought experiments fall by the wayside in the light of new knowledge, new experiences. That is just how it is, but this is no reason to abandon further experiments, or even refinements of the originals, be they scientific or thought.
I can think of no core philosophical principle which has been suggested might settle this matter, if in fact it needs to be settled. I am not convinced that it is possible to divorce thought experiments and physical experiments. Surely the process of observation, formation of an hypothesis, and the manner in which the hypothesis is to be verified, is in itself a thought experiment, but of course one which may be practically tested. The creation of quantum mechanics and relativity are highly dependent on the crucial role played by thought experiments. Contemporary philosophy, even more than the sciences, would be severely impoverished without them.
@Don Bird
Funny you should mention their role in the development of the “new physics”. I was thinking about TEs the other evening after finishing the Farmelo biography of Paul Dirac: a mathematician with a keen sense of visualisation.
On Jackson: by hypothesis she has all the physical facts; not all the physical facts as known at date of publication.
Re Andy Walsh:
I can’t find my copy of the Jackson paper just now and did in fact want to see exactly what he said about “knowing all the facts”
I am a bit puzzled here because would it not be a rash claim at any time that we know all the facts? How could you know that you knew all the facts? Scientific progress seems to conspire against this. Is it good science or good philosophy to hypothesize something which we know could never be the case?
The thought experiment is driven by the hypothesis that we know all the facts; it relies on this as a logical possibility rather than a realistic possibility. That’s why it’s a thought experiment. You can deny that we could ever know all the facts (in much the same way as you could deny the possibility of Parfit’s teleportation examples) but that’s a refusal to engage with the thought experiment, rather than an objection to it. Or so it might be claimed.
Mary and her release from confinement has been a matter of less than consuming interest, but the comments nudged me a bit.
I’d ask a different question, to wit: What is the relationship, if any, between explanatory, intellectual knowledge acquired indirectly via educational material and a perception acquired directly via sensual stimulation?
The way the experiment is typically framed we’re impelled to assume that Mary is expected to associate her knowledge with the red experience.
I tend to think that thought experiments in general won’t answer serious philosophical questions, but that they can function quite effectively to point out phenomena that stand in need of explanation (and perhaps for other uses). So while Jackson, on the basis of the Mary example, is not entitled to infer that materialism is false, he succeeds in showing that there is a challenge that the materialist must face: how are phenomenal states to be accounted for within a materialist framework?
I don’t believe that Rawls’ original position suffers in the way you say, though. It is a conceptual device that models constraints on moral reasoning that we already accept, in a way that, hopefully, helps us to see what we are committed to in the domain of social justice. Real agents, you and I, never go behind the veil at all. Real agents use this device to understand the commitments they actually have, as real, embodied, moral beings. It isn’t so much different than thinking, “What would I do if I were in her shoes?” Surely that way of thinking doesn’t undermine my agency, but is just a part of it.
[...] recently came across another post on the website, Talking Philosophy, that, at least at its outset, refers to a philosophical [...]
Compare “thought experiments” to any serious engineering task. An engineer can assess the situational environment, plan, analyze the plan for failures, redesign, and repeat this process until he is confident that the design will work. But in many, if not most, circumstances something will come up major or minor, that was not fully understood until the task was undertaken. Thought “experiments” should be viewed in the context that by their nature they are generally too impractical to ever be implemented. What one can learn from them must consider that you will never know what assumptions you may have made (and they are huge in the case of Mary here) that would give rise to such inconsistencies that would invalidate your “experiment”.
Andy, why couldn’t a brain in a vat imagine itself a brain in a vat, if it didn’t know it was a brain in a vat?
What is the difference between a brain in a vat and a brain in a skull, other than location? Mentation in both cases is surely identical in type.
Putnam’s argument is a defence against scepticism.
Thought (1): “I seem to be sitting watching Murray play Nadal at the Australian open but I could in fact be no more than a brain in a vat being scientifically manipulated into believing this. There may be no Murray. There may be no Nadal. There may be no tennis”
The thought seems perfectly coherent. What makes it coherent? The fact that it is meaningful. But if I really were a brain in a vat then I would lack a set of well formed concepts which would make it meaningful. The fact that it seems to make sense is what in the end refutes it.
Putnam’s argument assumes (or if you like defends) certain claims about meaning and content: including the claim that the content of a mental state T is external to T. So (@Don Bird) to state that “mentation in both cases is surely identical in type” is to make the mistake of believing that “meanings are in the head”: something Putnam rejects.
It’s a matter of internal honesty, integrity and coherence or following the argument wherever it might lead us. The images that we entertain bring a constellation of others in their train. Let’s call them a bound association. As they arise conflicts are noted. Is this conclusion necessary or arbitrary? I might choose to stress test a general principle by devising limit cases etc etc. General Principle: Keep it simple and stupid. Avoid complex set ups which are certain to have assumptions larded between the folds like butter in pastry. It’s the karate principle, focussing the force at a single point for maximum effect.
‘Jackson’s Mary’ leads you by the nose to the conclusion that you can have concepts without having the qualia that supposedly go with them. A much more secure demonstration might be had from Geach (Mental Acts) showing that it is a mistake to link the having of concepts and a specific experience. But can you have concepts without any concomitant (what does that mean?) experience? Can a blind person have the concept of colour or a specific colour?
Re Andy Walsh 26th Jan.
Putnam concluded his paper “‘meanings’ just ain’t in the head.” . This has always puzzled me. If some or all meanings are not in in the head where are they then? The world outside human experience does not do MEANING.
Tyler Burge subsequently argued in “Other Bodies” (1982) that the twins’ mental states are different: Oscar has the concept H2O, while Twin Oscar has the concept XYZ. Putnam has since expressed agreement with Burge’s interpretation of the thought experiment. (See Putnam’s introduction in Pessin and Goldberg 1996, xxi.) {Extracted from WIKI}
Whether a brain is in a vat or a skull all it ever receives from the world without is electrical irritation of nervous tissue. This is organised and deployed by the brain as necessary, particularly with respect to survival of the organism. I assume here that the vat is organised such that it replicates in the brain what we might call standard human experience i.e. the vat brain is not under the influence of a malevolent operator who is intent on sending it mad. Whatever the case madness or sanity, as Burge argues, mental states can differ from one person to another to which I would add also from scull to vat. Our interpretation of reality seems sufficient that we do not “bump into” each other too much in the world. In other words The mental states of each organism are peculiar to itself but there is apparently sufficient overlapping between organisms to permit success and progress in the environment. Not all mental states, wherever they originate, manage achieve this compatibility, which severely reduces the survival value of the organism in which they occur.
Andy Walsh says:- “But if I really were a brain in a vat then I would lack a set of well formed concepts” I am not sure what these well formed concepts are, or why the BIV would lack them.
[ A pity about Nadal; I did however have Murray down to win in three]
Thought experiments aside, the Mary example is of course contra materialism. It’s just one more variation of many on Cartesian dualism. I think Descartes did a good enough job years ago on selling the mind-body distinction. Enough clever up-dated versions.
Notice that a thought experiment is needed to go from “knowing all” about the physical (the brain), to the experiential (seeing the color red), but not the other way around: while seeing the red something (not a tomato, please!) the real experimenter can be detecting whatever relevant goings-on are happening in the brain. Somehow that’s not good enough for the dualist. Descartes lives on! Even though we know he’s dead.
I too don’t understand why a BIV would “lack a set of well-formed concepts” either. In the Matrix, Neo moves from the illusory world to the real world easily, with little re-adjustment. Brains are genetically programmed to be moulded by their environment in order for them to function in whatever environment they find themselves growing up in, so as Neo grew up in the Matrix, his vat-esque world must have provided him with enough sets of “well formed concepts” for his brain to seamlessly continue functioning just as well in reality as it had in the Matrix. In this scenario, I don’t see why the pre-enlightened Neo couldn’t have wondered whether he wasn’t just a brain in a vat, perhaps even after seeing (or being tricked into thinking he was seeing) a movie or reading a blog about it.
Well far be it for me, a through and through internalist re mental content, to step up and defend Putnam. For P being “well formed” is a matter of having the right sort of causal relations with the world and its community of linguistic competents. “The Meaning of Meaning” is a good one to get teeth into. Enjoy the trip to Twin Earth!!
One of the things she might have learned is how to shop for ripe tomatoes. (A spectrometer is a bit bulky to take to the market). Also how to participate meaningfully in human conversation about colour.
This is probably not a philosophical comment, but if not, why not?