A popular strategy in philosophy is to construct a certain sort of thought experiment I call an intuition pump...Intuition pumps are cunningly designed to focus the reader's attention on "the important" features, and to deflect the reader from bogging down in hard-to-follow details. There is nothing wrong with this in principle. Indeed one of philosophy's highest callings is finding ways of helping people see the forest and not just the trees. But intuition pumps are often abused, though seldom deliberately.In his 1991 book Consciousness Explained, however, Dennett deploys the metaphor polemically:
The most influential thought experiments in recent philosophy of mind have all involved inviting the audience to imagine some specially contrived or stipulated state of affairs, and then—without properly checking to see if this feat of imagination has actually been accomplished—inviting the audience to “notice” various consequences in the fantasy. These “intuition pumps,” as I call them, are often fiendishly clever devices. They deserve their fame if only for their seductiveness.
 |
| The pains and pleasures of intuition pumps. |
Before examining the elements of Nozick's intuition pump, however, it is worth briefly attending to the "intuitions" that are being "pumped" by these metaphor(ical) machines. What are these “intuitions?" The term "intuition" has a long history in philosophy, and we need not go into it here. For present purposes it is enough to note that what philosophers are frequently trying to obtain with their production of/appeal to "intuitions" are non-inferential propositions from which they can draw subsequent inferences. Intuitions are sought out so that they can be made to form the foundational bedrock upon which the rest of a philosophical apparatus can be built.
Again and again one finds in the literature that we have a "prima facie case for saying that X," or simply "here we find that we have the powerful intuition that X," and so on. In Nozick's case, the intuition he wants pumped is an immediate sense that "something matters to us in addition to experience," an intuition we obtain by "imagining an experience machine and then realizing that we would not use it." If Nozick can get us to agree that when imagining an experience machine that we do indeed have the intuition that he predicts we will have--that we do not want to use the experience machine, then he can use this intuited assent to build his case that we want something more than "how our experiences feel from the inside."
To analyze an intuition pump is to try trace the origins of whatever "realizations" the pump is trying to push. To whatever degree we are able to re-calibrate the mechanisms that pump these "realizations," we may find that the intuitions the pump had previously pushed "into" us are no longer as clear as we had previously supposed. Let us consider then, the elements of Nozick’s intuition pump, the components of his experience machine--and some of the ways in which they mislead us.
1. The very idea of an experience machine
Before digging into the individual elements of Nozick's experience machine it worthwhile to briefly rehearse the fact that we can conceive of such a thing at all, with the concomitant consideration that how we think about experience will impact how we evaluate experience inducing machines. In part one I detailed some progenitors to our modern conception of an experience machine. In the present age of radical technological acceleration, the notion of an experience machine has acquired additional plausibility due to overwhelming evidence that the bulk (if not all) of what we call consciousness is an "output" of the brain. While Plato's cave must content itself with imagery that amounts to little more than a modern movie theater (though this in itself is remarkable, that it took us some 2,000 years to actually MAKE Plato's cave) and Descartes marveled at the simulacra produced by dream states (postulating that an "evil demon" might serve as the diabolical host of our experience), we can now consider with all seriousness the possibility that experiences could be injected by a literal machine directly into the "mind" via the brain, skipping contact with the ears and eyes and other organs of sense, creating a world that is only known "from the inside." Even more radically, "strong AI" proponents suggest that we might be able to create machines that themselves have experience. The sense that experience machines (of whatever type) might soon be literally possible, as opposed to merely metaphorical flourishes, gives additional cultural weight to "brain in a vat" thought experiments, of which Nozick's experience machine can be considered an exemplar. Typically such thought experiments are focused on epistemological questions: how can we know that we aren’t already brains in a vat? Such radical skepticism has managed to force the expenditure of much ink in defense of our unvatted natures, but Nozick’s argument can be considered primarily a normative attack on any desires we might have to be so envatted.
Despite these recent developments we should be wary of allowing the increasing literality of the experience machine metaphor to obscure our awareness of the non-literal conceptual schemes that enable us to consider them in the first place. The structures of Plato’s cave and Descartes evil demon metaphors remain relevant to our considerations because we are still holding on (perhaps erroneously) to key concepts embedded in those metaphors. As Dennett notes at the end of Consciousness Explained:
I haven’t replaced a metaphorical theory, the Cartesian Theory, with a nonmetaphoircal (“literal, scientific”) theory. All I have done, really, is to replace one family of metaphors and images with another, trading in the Theater,the Witness the Central Meaner, the Figment, for Software, Virtual Machines,Multiple Drafts, a Pandemonium of Homunculi. It’s just a war of metaphors, you say—but metaphors are not “just” metaphors; metaphors are the tools of thought. No one can think about consciousness without them, so it is important to equip yourself with the best tools available. Look what we have built with our tools. Could you have imagined it without them?Indeed we cannot do much imagining without our metaphors, and this is precisely the reason we must be so careful when constructing intuition pumps that “help” our imaginations produce intuitions. In the case of the experience machine metaphor, it should be noted that while Dennett has done a great deal to destabilize and/or re-imagine the Cartesian Theatre and the Central Meaner, he has done little to soften our sense that “experience” is rooted in a machine. On the contrary, he has intensified it. While previously the experience machine was either an apparatus or an agent that produced experiences for, and delivered experiences to an observer; functionalist approaches tend to imagine experience is the output of a virtual machine, software running on the wetware of the brain.
A full analysis of how these various metaphors interact with our sense of what an experience machine could or could not do is obviously pertinent to our considerations here, but will have to be set aside due to the scope and subtlety of addressing the questions in play. For now simply note that the various metaphors of “machine,” “apparatus,” “theater,” and “agent” are remarkably difficult to shake, and that it is unlikely that the addition of “Software, Virtual Machines, Multiple Drafts, [and] a Pandemonium of Homunculi,” to our stock of metaphors will in and of itself provide an answer the question: “do we want something more than experiences felt from the inside?” For even if we succeed in banishing “the inside” into a pandemonium of homunculi, we will not suddenly find that we have no experiences left to introduce via an experience machine. Rather, we will find we are the experience machine that is busy producing experiences. The question may then simply become: "Do we want to be something more than experience production machines?"
2. How do the experiences get injected?
Nozick suggests that we imagine the experience machine as tank in which we float, with "electrodes attached to [our] brain." This image is clearly intended to evoke "brain in a vat" imagery, but as stage setting for an investigation into our hedonistic preferences it is needlessly biased. The question at hand is: Is there something more important to us than “how our experiences feel from the inside?" Given that we have no idea how exactly we would implement such technology, the notion of “floating in a tank," seems an arbitrary conceit, one that Nozick explicitly frames in a pejorative manner, stating “someone floating in a tank is an indeterminate blob." Clearly this is not a charitable vision of how an experience machine might be implemented. Given that the possibility of a true experience machine is distant at best, and given that we have no idea how full blown “experience technology" will work, we need not generate invidious descriptions of "blobs floating in tanks," imagery akin that found in the nightmarish film "Altered States". The reader can of course imagine any number of contraptions, and popular culture can furnish many more: the fleshy consoles controllers of Cronenberg's "Existenz," the gruesome vats and robotic IV injectors of "The Matrix," the sleek modern recliner chair of Joss Wheaton’s "Dollhouse," The MRI-like scanner of "The 13th Floor" and so on. To these we could even add drugs like Phillip K Dick’s Chew-Z.
However the experience machine is imagined, what is crucial is that we recognize the influence that the mode of experience delivery can have on our evaluation of the experience machine itself. We might not be attracted to becoming "indeterminate blobs floating in a tank." We might be more inclined to take a pill, or lie down on a comfortable recliner have have neural nets attached to our foreheads. Any experience machine thought experiment that really wants to test just our attachment to "reality," ought to remain as bland as possible in its description of experince injection. This is certainly a situation in which the medium can infect the message. Any serious test of intuitive responses to the experience machine needs to allow for a variety of experience machine injection mechanisms--in order test for responses not only to the experience machine itself but also how the experiences get injected.
In a certain sense, though, the question of how the experiences are injected is a purely aesthetic exterior coating, given that our central focus is the quality of the experiences "on the inside." Indeed, “the outside” becomes even more irrelevant if we to the experience machine for life.
3. How much time must be committed to the experience machine?
Nozick at one point seems to suggest that our commitment to the experience machine must be lifelong. He writes: "Should you plug into this machine for life, preprogramming your life's experiences?" Strangely he seems to back off this radical commitment a sentence or two later when he suggests that re-programming sessions will be allowed: "After two years have passed, you will have ten minutes or ten hours out of the tank, to select the experiences of your next two years." Nozick’s suggestions here seem to contradict each other. First, why even bother to add this intermittent visit to the real world? And if one is going to allow reprogramming sessions outside “the tank” why stop there? Why not allow a visit to family and friends? In other words, why isn’t the question "should you plug into this machine for 2 years, and then go out and have some cake and cookies?" For that matter, why isn’t the question "should you plug into the experience machine for 2 days, 2 hours, or 2 minutes?”
This confusion significantly impacts the quality of the intuitions that Nozick's thought experiment pumps as it is likely that the question of whether or not one is committing to the experience machine for life will heavily influence any decision to hook up to the experience machine.
On the one hand, a hardline inquiry testing whether or not we care about our connection to "real" reality would seem to favor an experiment that tests a more absolute commitment. For if our goal is to discover whether or not there is anything that matters to us other than how things “feel from the inside,” we cannot admit to our thought experiment anything but things felt from the inside. If we are given a safety valve, an “out” from our time in the experience machine, then we will be able to have our experiences and eat them too. Whatever it is that we want other than a diet of experiences felt “from the inside,” a thought experiment that offers us the ability to choose how much time we spend in the experience machine will surely give us the opportunity to imagine this balance. If people were given the opportunity to pick and choose when to don and when to remove the experience machine apparatus, it seems likely we would see lines forming around the block in anticipation of “experience adventures” doled out in the time increment of the purchaser’s choice.
On the other hand, it is not entirely clear that such a lifelong commitment to the experience machine is nearly as restrictive as it might first appear. I submit that Nozick's description of time spent outside the experience machine is a technical superfluity--the purpose of which is the accentuate the sense that time within the experience machine is "preprogrammed." For there is no clear reason that the machine must be reprogrammed from "outside" of its experiential matrix. We could, for example, simply allow for a mechanism that permits the next slew of experiences to be adjusted from within the experience machine itself. Why add the image of waking up in in a tank of water from which one must emerge for 10 min to select the next batch of experiences? Why is it necessary to exit the experience machine in order to alter the experiences it produces? One we make this adjustment, not only can we dispense with the “indeterminate blob” floating in the tank of water, we can dispense with the 2 year "reprogramming interval" as well. If we are considering what it would mean to commit to a life lived only “on the inside,” then we ought to stipulate that no emergence from the experience machine is required to adjust the experiences it outputs to our "interior." Whatever experiences we imagine the experience machine offering, we should also imagine this crucial additional experience: the ability to experience ourselves altering the experiences we are having. But this change suggests that an experience machine might offer us an unparalleled experience of freedom.
I believe that it is precisely a sense of a loss of freedom that is illegitimately being pumped by Nozick's experience machine. Whether we imagine the ability to freely choose whatever amount of time we spend inside the experience machine, or whether we imagine being able to re-map the experiences we are having from inside the experience machine, we still wind up with an intuition of freedom that is substantially greater than that afforded by the awkward confusion of "lifelong commitment" with intermittent "preprogramming" sessions.
4. What does it mean to "preprogram" our experiences?
Immediately after asking us if we are willing to commit to the experience machine for life, Nozick suggests that use of the experience machine "preprograms" our experiences. "Should you plug into this machine for life, preprogramming your life’s experiences?" This question flies by so quickly, that the intuition it pumps is easily overlooked, a seemingly minor component of Nozick's experience machine. But we should think carefully about the assumptions made by Nozick’s question, for the notion that experiences can be “preprogrammed” goes right to the heart of one of our most fundamental concerns: the question of whether or not we have free will.
It is all too easy when reading Nozick’s thought experiment to accept that all of the experiences produced by the experience machine will be determined, that the possibility of choice will be foreclosed once the experience “program” is engaged. But framed in this manner, the question does not assist but misleads our imaginations. Nozick may as well have asked: would you like to enter an environment in which you have no freedom, in which everything that happens to you is preordained? It is easy when considering this question to assume that we are in a non-programmed environment right now, that we currently have free will, but that upon entering the experience machine that all our experiences will become "preset." Thus Nozick has given us not only the anxiety that we will lose contact with a "deeper" reality, but also lose our precious ability to determine our future (excepting, of course, our initial choice of the pre-programmed experience, selected from the experience machine’s library of experiences).
But do we currently have free will? Are there not analogs in daily life of “experience programming?” Are there not many activities we engage in precisely so that we might have experiences that we are passive recipients of? Obvious candidates are things like TV and movies, our proto experience machines. But the question is more broadly applicable. Do skydivers "preprogram" a rush of adrenaline when they jump out of airplanes? Is the psychiatrist trying to "preprogram” her patient’s future experiences when subscribing a particular anti-depressant (from her library of antidepressants)? Is the patient who visits the psychiatrist requesting an opportunity to “preprogram” themselves or "reprogram themselves? Is there a difference? Do our genetics create "program limitations" on the range of possible experiences we might have? Are the covers of supermarket magazines “programmed” to get some segment of the populace to pick them up? When members of the demographic in question pickup said magazines are they doing so because of "internal programming?"
All of these questions would be interesting to pursue. But we need not pursue them here. It is enough to note that it is not clear where we should draw the line when it comes to our real world practices and their relation to free will. It is not clear to what degree we are either “programmed” or engaged in “preprogramming” as we go about our daily activities. Certainly it is not obvious that we are free now in a way that we would not be upon entering an experience machine, for we have neither settled what the necessary and sufficient conditions of real world freedom are, nor which of those conditions would or would not be transferred over to the environment offered by the experience machine.
Indeed, Nozick has only given us the barest sketch of how the experiences “inside” the experience machine would unfold. He writes: "Superduper neuropsychologists could stimulate your brain so that you would think and feel you were writing a great novel, or making a friend, or reading an interesting book." But here we are left with many questions. What would it mean to “have the experience of writing a great novel?” Would you be able to experience what it was like to have written Hamlet? Could you be Shakespeare as he wrote the play? Or is it that in the experience machine one believes that one is writing a great novel, but that no such novel is actually produced at the end of the experience? If one returned to the real world after “writing a great novel” inside the experience machine, would the memory of the experience remain intact? Would one remember having written this novel as though it were one’s own? Would there indeed be an extant great novel produced as a result of this process? If it was a great and original novel, would that mean that the machine had written it "for" you, and then produced an illusion in your mind that you had written it on your own? Is the preceding question coherent?
The lack of obvious answers to the preceding questions suggests that we should not cavalierly accept Nozick’s suggestion that the machine would “preprogram our life’s experiences.” It is enough to merely ask: “would you plug into this machine for life, so that it was the exclusive source of your experiences?” So framed, the question loses most of its negative connotations, for it now only asks of us that we choose between two sources of experience: the atoms and void of the real world, or the ones and zeros of the experience machine.
It is particularly crucial to clean out this portion of the intuition pump given how freedom focused our culture is, a culture where the terms "program" and “preprogram” are likely to invoke deep anxieties. It is one thing to program a rocket ship and send it to the moon. That is a great accomplishment. It is another thing to preprogram people. To engineer a robot is a technological feat; to engineer human beings is to threaten dystopic nightmares.
5. What happens to everyone else while you are on "the inside?"
It flashes by quite quickly, but Nozick does offer some consolation for our friends and family--those whom we will lose and who will also lose us--should we leave the "real world" for the temptation of the experience machine. "Others can also plug in to have the experiences they want, so there's no need to stay unplugged to serve them. (Ignore such problems as who will service the machines if everyone plugs in)."
Never mind that our human relations are reduced to the "serving" of "others," the question of "what happens to everyone else while you are on the inside?" is surely an important one. For most humans, personal relationships form a substantial portion of what makes life worth living. Failure to spell out clearly what will come of one's closest personal associations upon hooking up to the experience machine allows too much room for anxieties to arise about the things many of us care about most: our favorite human beings. To see this one need only imagine to what degree the emotional tenor of the experience machine would change if it were possible to hook up to the experience machine in groups, such that the various experiences provided by the experience machine could be shared with friends and family. Or even to allow for friends and family to visit the experience machine enthusiast by occasionally "dropping into" the experience machine environment.
It seems quite plausible then that the question "will we lose contact with those we care about most upon connecting with the experience machine?" may well be obscuring the question of whether or not we care about "losing contact with 'real' experiences."
Re-calibrating the experience machine
Nozick believes that his experience machine shows that "we learn that something matters to us in addition to experience by imagining an experience machine and then realizing that we would not use it." But as we have seen there are far too many variables introduced by the experience machine to give us a clear sense of what this "something in addition to experience" might be.
First, there is the question of how the experiences are injected into our "interior." There is no reason to imagine the experience machine in the grotesque way that Nozick does. If what we are testing is whether or not "something matters to us in addition to experience" we should strive to test just that intuition and not also our intuitions about being indeterminate blobs in tanks, particularly when it is not clear that it is necessary that we "become" the latter.
Second, there is the question of how much time must be spent in the experience machine. Nozick's writing is ambiguous on this point. Making the commitment to the experience machine "for life" is a good way to test whether or not we need "something that matters to us in addition" to just a life inside the experience machine. But Nozick is substantially ambiguous about what the precise features of this lifelong commitment will be. He suggests that there will be "break" from the experience machine of up to 10 hours during which we can re-program it. He does not establish the ground rules for contact with other people during our life long commitment to the experience machine. Finally, he does not make it clear how much control we will have over the experience machine over time.
Third, and closely related to the question of time, is the question of what it means to "preprogram" our experiences. Nozick seems to suggest that the experiences produced by the experience machine are "prepackaged," that we are selecting the experiences from a "large library" of experiences. But there are many issues here. It is not clear how long any given experience must be, before it is subsequently modified by the user of the experience machine. It is not clear how much control the experience machine user has over any given experience. Most importantly, we do not have a clear idea of how much the metaphor of "programming" relates to our real lives. People regularly seek out and "consume" experiences during the normal course of life. Recreational drugs, television and movies, even exercise and meditation can induce altered states of consciousness. Can we clearly delineate between these forms of "experience induction" and Nozick's experience machine?
Fourth, and quite crucially, there is the question of what happens to the people we care about most once we enter the experience machine? Can they come with us? Must we go alone? Can they visit us inside our happy dream? Might it not be the case that sum total of what we want "in addition to our own experiences" is contact with other people? Nozick's thought experiment does little to help us settle this question.
Finally, there ambiguous nature of experience itself. While questions about "the nature of experience" are rich enough that they offer an entire separate line of inquiry, it is important to keep in mind that any project that mediates on our feelings about experience machines rest on a host of metaphors and presuppositions about how experience works in the first place.
In the next section we will examine some of these components more closely, with particular emphasis on the question of time, in order to see how unclear it is that there is anything that matters to us "in addition" to experience.
No comments:
Post a Comment