Difficulty: What the fuck
This is Part 3 of a series on Robert Nozick’s Experience Machine Thought Experiment (EMTE), so make sure you first read Part 1 and Part 2. As I noted earlier, the philosophical discussion here is a simplified and slightly profane version of the Internet Encyclopedia of Philosophy’s article “The Experience Machine.”
So just when you thought things were complicated enough, there are more objections to Nozick’s EMTE argument. Here, we cover methodological challenges, the expertise objection, the experience pill thought experiment (an offshoot of EMTE), and newer variations of EMTE, before we conclude with a brief note on what the hell all this suggests.
Methodological challenges
First, what the hell does “methodological” mean? Here, “methodology” means the way a thought experiment or psychological study was set up. So, if it’s okay to sacrifice a bit of linguistic precision: methodological challenges are simply objections to how shit is done.
The first methodological challenge is raised against the previously mentioned psychological studies, which aim to see how participants make decisions when confronted with EMTE or EMTE-like scenarios. Specifically, the objection is that these empirical studies are not representative of the general population: they tested only students in Anglo-Saxon universities, so their results are not enough for us to conclude anything about how the rest of the English-speaking world would make decisions under EMTE, let alone how most people, globally, would make such decisions.
The flaw in this objection is that it misses the point. The goal of EMTE, REM, and stranger NSQ is not to understand what humans think in general—that’s a sociological or anthropological thing. Instead, the goal is to challenge how philosophers’ use of intuitions (like “I wouldn’t plug in”) to argue that we value reality, or that hedonism is false, might be unreliable because of our biases. In other words, we don’t need to know what the whole human race thinks in order to weaken Nozick’s abductive argument. It’s enough that these studies produce inconsistent results so that the old line of thought—”Oh, an overwhelming majority of people wouldn’t plug in, so they must value reality, and therefore hedonism is false”—starts to look much shakier.
The second methodological challenge is raised against the fact that these psychological EMTE studies did not set up hypothetical scenarios as real decision-making situations, but as outlandish scenarios: wouldn’t it be better, the objection goes, if the studies had made participants make decisions about more “real-world” situations rather than some fantastical sci-fi stuff?
The answer is “probably no.” In fact, the fact that EMTE is so different from real life may actually be a strength of the setup, not a weakness. Why? Because we are even more likely to exercise biased decision making when we are confronted with real-world situations. So the point of EMTE and its variations is not to see how we’d act if we were actually standing in front of a machine. The point is to examine our judgments in a setting that may be less driven by immediate emotions and pressures, and therefore better suited to philosophical reflection.
The expertise objection
The expertise objection basically says: “Hey, we should be using only academic philosophers’ responses in experimental philosophy because they are ‘better’ at thinking through thought experiments.” And if we believe that, we’d also believe that non-philosophers’ responses to these thought experiments are not meaningful for philosophy. This is a common challenge not only to EMTE, but to experimental philosophy in general.
The flaw with this objection is that empirical evidence shows that, for both EMTE and experimental philosophy in general, philosophers are not way better at thinking things through. When asked to respond to different versions of EMTE, philosophers gave inconsistent answers, and their answers were only slightly more consistent than those of non-philosophers. In fact, philosophers are around equally susceptible to imaginative failures as non-philosophers; those two populations disregard stipulations of EMTE at approximately the same rate.
Further, if philosophers really want to claim their status as the hot shits of thinking, then, if we are to remain democratic about things, the burden of proof lies on philosophers: philosophers must prove that they are better thinkers than non-philosophers, not the other way around. But the evidence, as noted before, does not support that claim.
Indeed, philosophers might be susceptible to other biases coming from within academic philosophy itself. For instance, given that the mainstream position in philosophy is that one should not enter EM, if a philosopher participant does not specialize in EMTE and thus relies on more muddled thinking, then that philosopher might more easily default to the anti-plug-in answer simply because that is what philosophers are “supposed” to say.
The experience pill
Then there’s the experience pill thought experiment, an offshoot of EMTE: would you take a pill that enhances your pleasure, but that does not distort reality like psychedelic drugs do? With this modification, pro-pleasure decisions increased from 29% to 53%. This might be because the experience pill thought experiment presents a much more familiar scenario and does not require pro-pleasure decision makers to abandon reality.
This points to an important difference between EM and the experience pill: while interesting, the experience pill shifts the question. It is no longer really about reality versus pleasure. Rather, it helps us analyze whether people prefer affective appropriateness—that is, having feelings that fit the situation—or pleasure.
A related issue: as is the case for much of experimental psychology and philosophy, changes in how you word a thought experiment often change participants’ decisions.
For instance, if I told you to “watch this video of a ‘car crash‘” and then asked you to estimate how fast the car was going, you’d probably be more inclined to say that the car was going much faster than if I had worded the question using “car accident” or “fender bender.”
So, because EMTE is just as susceptible to inconsistent responses due to changes in wording, and the wording of EMTE varies widely across studies, it’s difficult to know just how much of the inconsistencies are due to mere wording variations rather than to the deeper biases we’ve already discussed.
New EMTEs
In the past two decades, philosophers have come up with new versions of EMTE so that a new and stronger argument can be made against hedonism. Specifically, one version of EMTE attempts to eliminate worries that EM would malfunction (imaginative resistance) and the status quo bias. In this version, you are asked to decide between two lives:
(1) Life A is “pleasant, rich, full, autonomously chosen, involves writing a great novel, making important scientific discoveries, as well as virtues such as courage, wittiness, and love.”
(2) Life B is identical to Life A except for the fact that B is lived inside EM.
This is called the experientially identical lifetime comparison.
The argument goes that this new version of EMTE isolates the variable that actually matters: our relationship to reality. By conducting the thought experiment this way, we can more easily see whether people care about reality. And because it seems that most people would not want to live Life B (the life in EM), this suggests that we do care about reality, which pushes the argument closer to the conclusion that reality might be an intrinsic prudential value.
Rach Cosker-Rowland, in her 2017 paper “Our Intuitions about the Experience Machine,” presented a version of EMTE where the participant decides which of two hedonically equal lives a stranger should choose. But the IEP article this post is based on (following Buscicchi) claims that Rowland’s study contains a “macroscopic” methodological mistake that “severely undermines the significance of Rowland’s study.” The alleged mistake is that she did not provide participants with the option of answering that the two lives have equal value.
And then there’s Shelly Kagan’s deceived businessman thought experiment, which, though far less famous in philosophy, is considered by some within philosophy circles to provide the strongest case against hedonism, partly because it doesn’t require participants to imagine a life of virtual reality. Here it goes:
Deceived businessman thought experiment
Imagine that in life A, a successful businessman lives a happy life: his family loves him; his community and colleagues respect him.
In life B, the businessman lives an equally happy and pleasurable life, but he is completely deceived: though everyone around him acts the same way, they do so for their personal gain.
Hedonism would hold that these two lives are equal in value, but our intuition is that life A is better than life B.
But of course there are objections. Some philosophers argue that our intuition in favor of life A is being driven by moral considerations rather than purely prudential ones. We value life A more because, unlike life B, it contributes to the world. But again, moral questions are irrelevant and violate the stipulations of the thought experiment, which instructs us to consider only prudential (the “what’s good for me“) reasons.
Another possible flaw of this thought experiment is the freebie problem. First, it is irrational to have 100% confidence in the truth of hedonism. So if we are less than 100% sure that hedonism is correct, the rational decision is to prefer life A, which has a greater-than-0% chance of providing a prudentially more valuable life, since it gives us the freebie of life without all that deception. (And if we think back to Rowland’s study, which did not provide participants with the option of answering that the two lives are equally valuable, we see how forced dichotomies like that are even less helpful for tracking whether we value reality.)
Also, the original EMTE is arguably stronger than the deceived businessman thought experiment because the former requires participants to sacrifice one of two values on a much greater scale: to live a fake life in EMTE is not just to live a life in which you are deceived by others; you’d have to go for full-blown virtual reality for the rest of your life. And if you chose to live a real life, that would mean giving up a huge amount of pleasure that you could otherwise have had. By requiring participants to make great sacrifices like these, EMTE allows us to more clearly see what participants actually value. And compared to the original EMTE, the sacrifices required in the deceived businessman thought experiment are tiny.
Lastly, Kelly Inglis asks us to imagine, in her 2021 paper “The Universal Pure Pleasure Machine: Suicide or Nirvana?”, a universal pure pleasure machine (UPPM) that gives every sentient being in the world a constant heroin-like bliss. Then she asked her participants whether this is “a good future that we should desire to achieve?”, to which only 5.3% of participants responded positively. Inglis concluded from these results that UPPM can disprove hedonism, but not all philosophers are that confident about its implications. For one thing, moral judgments tend to be universal (for example, everyone should not commit murder). For another, UPPM might still make us susceptible to imaginative failures: people might disregard the stipulations and see the machine, which has no virtual reality, as boring (imaginative resistance), or go beyond the stipulations and think that its constant heroin-like high is disgusting (overactive imagination).
So WTF is the conclusion?
This is where I pick, more or less arbitrarily, what I take to be the most salient points of this series on EMTE:
Nozick’s EMTE went through two major stages. It began with the publication of Anarchy, State, and Utopia in 1974. Up until around 2010, an overwhelming majority of philosophers accepted EMTE as support for a knock-down argument against hedonism. Then, beginning in roughly the 2010s, a shit ton of philosophers came up with objections that attempt to weaken EMTE, making us question whether EMTE provides a good argument against hedonism at all. Our biases might have affected our judgments, the scenario might be too far-fetched, and our intuitions might not be tracking value as cleanly as people once thought.
This second phase is then marked by philosophers who devised different variants of EMTE, like Kagan’s deceived businessman thought experiment, in their attempt to revive EMTE’s goal of disproving hedonism.
So where does that leave us? The experience machine used to look like a decisive argument against hedonism. Now it looks much more like a messy intuition pump, one that reveals as much about our psychological biases and philosophical methods as it does about what actually makes life go well.
At this point, the debate is no longer just about whether hedonism is true. It’s also about whether our intuitions, especially about weird, sci-fi scenarios, are reliable enough to settle philosophical questions in the first place.
And that’s probably the most important lesson of all this. EMTE is still interesting as hell. It may still pressure hedonism in serious ways. But it is no longer the philosophical slam dunk that many people once took it to be.
Philosophy without Bullshit 