The Philosophy of “Severance”: Does an “Innie” die if the “Outie” Retires? Derek Parfit's Response

Or, an evangelical philosopher and secular utilitarian walk into a bar and watch Apple+’s “Severance”…

Still from “Severance,” Apple TV, edited and repurposed

—Read time, 11 minutes (10 if you’d had some coffee)—

Severance, the hit Apple+ TV show, has its hooks in my wife and me. We only needed a free week-long trial to watch the 18 or so episodes to catch up (the season finale is next week).

The psychological thriller follows four office workers who are “severed.” Their work memories become completely separate from their personal memories. The version of themselves that’s working has no memories of their outside self—they’re born anew. The outside self doesn’t remember their time at work. Ahh, the ultimate work/life balance! Nothing could possibly go awry.

Effectively, this separation creates two streams of consciousness, identities, and story lines. The “innie” (the worker) and the “outie” (the one with a personal life), never communicate and are forbidden from making contact. You can’t even write a note to yourself. The “outie” worker holds all the power, as they decide whether to continue working or not, despite never experiencing the job. The innie effectively lives in purgatory, as they “wake up” at work the moment they leave for the day—the rest is blocked from their memory.

“Severance,” Apple TV

The corporate baddie behind this twisted work place, Lumon Industries, is all smiles, insufferable office jargon, cult worship of their founders, and bureaucratic doublespeak, to the point of hilarity. The fake niceties and mind-numbing LinkedIn-speak probably hits a little too close to home for some folks.

How can these “innie” workers escape their workplace hell? What else is going on at Lumon? The best way to describe what happens in the following episodes it’s a little like Lost, but better, and it’s set in an office space. (Reminiscent of the backrooms, for internet natives out there.)

Suffice it to say, the show is engrossing, taking you do endless hallways of twists and turns, prying philosophical and ethical dilemmas through characters you genuinely care about. In addition to being existential, moving, and unsettling, Severance also happens to be riotously funny.

Does the innie in “Severance” survive?

Here’s one of the central questions of the show: Are the innie and outie the same person? Do they each deserve individual rights? What happens if an innie falls in love with another innie when their outie is married?

Here’s the one we’ll focus on. If the innie is “retired,” and the outie doesn’t keep any of the innie’s memories, does the innie survive? In other words, if the outie quits, and never comes into work, does the innie, with their exclusive, unique memories and stories, die, as they never experience anything again?

Today, we’ll look at a preeminent 20-21st century philosopher, Derek Parfit.¹ He has some radical ideas put forward in his 1971 paper, “Personal Identity” which seems well-suited to answering this question.²

Out of my wrestling with this paper, I’ve been working on a concept called a “spiritual continuum.” But before we attend to the theological reflections, let’s deal with this cheeky paper by Parfit, and answer your burning Severance questions.

What if I were split in half?

To answer one sci-fi inspired question, we’ll need to look at another one first.

Surprisingly, people can survive with one half of their brain. A hemispherotomy severs one half, or hemisphere, of the brain from the other. Now, let’s assume we could also do brain transplants.

Imagine a case where you were split into two people. A surgeon severs the halves of your brain, and puts each half in a body; a freaky, but not uncommon sci-fi thought experiment. Say the operation is a resounding success, and each half of your brain becomes fully healthy and functional. After the operation, you wake up—or do you?

Here’s the question Parfit poses: Does the original you survive?

Personal identity is one-to-one

Personal identity is binary, black and white, one to one. I’m Mark, not anybody else. Nobody else is me.

Ordinarily, Parfit thinks, this kind of language works well. However, there are some problem cases, like the one I gave a moment ago. If you were to split into two, would you survive? Parfit thinks there are three options for the person who gets a double brain transplant.

1. Pre-operation person does not survive.

2. Pre-operation person survives as one of the two people.

3. Pre-operation person survives as both people.

It seems (1) and (2) are implausible. The first option seems unlikely, because we say the pre-operation person survives in the case of one hemispherotomy, so why not a double hemispherotomy plus transplant? As he writes, “How could a double success be a failure?”

Second, against (2), there isn’t a good criterion for choosing one versus the other resulting two. (For the purposes of the thought experiment, the two resulting people are essentially clones.)

Finally, according to Parfit, (3) is the most plausible, but is problematic because identity implies a binary—black and white, all or nothing. As he says, identity is a “one-to-one” relation. I need to be different from everyone else to be an I, right?

Personal identity isn’t the foundation

Parfit goes with an alternative to all three. A bonkers option. We simply “give up the language of identity. We can suggest that I survive as two different people without implying that I am these people.”

In other words, he says, there is no correct answer about which of the post operation people are the original person, given the concept of personal identity. How is this possible?

Consider national identities, like England. We want to say England is a nation. It has an identity. It’s a one-to-one relationship. England is not Brazil, or the US, or any other country. England is England.

However, some cases, like the question “Did England become a different country after 1066?” is a question about identity without a correct answer. It’s too gray an area, because what matters to us about what makes a nation, a nation doesn’t definitively decide yes or no, either way.

By this, Parfit does not mean we need to give up talk of identity in our day-to-day lives. Rather, he wants to show how the concept of “identity” points to something deeper, more important, and more rational.

What we really care about, what’s underneath the concept of identity, is the notion of a “psychological continuum.” Let’s unpack this idea next.

The psychological continuum is what’s really important

According to Parfit, what’s important to us is really a matter of degree.

That’s because, what really matters to us, is clusters of psychological connectedness. For example, it doesn’t matter that Mark Legg=Mark Legg, but that I have a connected bundle of memories, values, intentions, skills, talents, ideas, emotions, beliefs, and more, that string together—maybe a bit like a web. These are all a matter of spectrum, according to him, not a matter of black and white distinction.

This is why, for example, someone might say, “I’m a different person now,” referring to their past self. Their values, emotions, beliefs, and more, have all changed enough that they feel like a different person altogether—despite having the same body. Indeed, if I lost all my memories, beliefs, values, ideas, etc., then wouldn’t I no longer be me?

As long as the psychological connections form a causal continuum, we describe them as continuous and threaded together, giving some notion of identity. Here’s a diagram he uses that’s supposed to help.

What in the Sam h*ck is this monstrosity? Here’s the explanation. Buckle up.

Remember the case of a person splitting? “A” represents an individual before they split into two people. B+1 and B+2 are the split people of A, so they’re “continuous” with A. Say B+1 splits again, as does B+2. We could imagine that each descendant is still continuous, but less connected to, A as they get more and more removed. So, B+6 may only have glimmers of the same personality, values, thoughts, ideas, memories, etc., because each split loses a bit of the original. In this way, each individual could be continuous with many descendants, if, in some freak scenario, they split, sci-fi style.

Here’s another diagram that might help.

Here, we can see the definition of psychological continuity play out. Each line represents a continuous line of psychological connectedness which exists in a casual relation. As long as we’re casualty continuous and maintain some psychological connectedness, one from the next, we are on the same spectrum. So, Parfit thinks, in a real way, we survive on in our descendants. Each hill in this line is two people, they overlap some, like we do with our children. As time goes on, they become more distinct, just as we become different from our past selves as time goes on.

Here’s a way to understand it:

A → A+1 (me tomorrow) … → A+50 (me in a decade)… → A+100 (my son) → A+1,500 (my great-grandson)… etc.

Whoa. Mind blown yet?

Nerdy Philosophy for a second

This claim is pretty radical. Let’s lay out his proposal in a rigorous, philosophical way. (If you don’t follow this, it’s okay—I’m just doing due diligence here. Read his paper to get the logic more fleshed out.) Here’s what I take to be the structure of his argument:

1. It is not a logical truth that “I” need to have my own memories or intentions (or any other psychological concepts).

   1. He argues for this first point quite a bit in the paper.

2. We can coherently describe two or more people “fusing” together and individuals “splitting” into multiple people.

3. Fusion and splitting are not clearly either survival or failure to survive.

4. Because fusion and splitting make use of spectrums of psychological connectedness and continuum (which can be described objectively), we can describe people’s survival in terms of a grayish spectrum—not black and white.

5. Therefore, what really matters in cases of survival is a spectrum of psychological connectedness, not personal identity.

So, when can we use the concept of personal identity? In most cases of day-to-day life. We could use the black and white concept of personal identity in this way: “X and Y are the same person if they are psychologically continuous and there is no person who is contemporary with either and psychologically continuous with the other.” In other words, it applies to about 100% of cases, which is why we use personal identity so often, with so much confidence.

What about my relationship to myself, as an “I?” Well, he says, “the word ‘I’ can be used to imply the greatest degree of psychological connectedness” (emphasis added).

Mind-blown now?

A handful of critiques—what about consciousness?

In my estimation, Parfit does an excellent job of dismantling the notion of one-to-one individual identity. On the surface, this could be taken as “no man is an island” or the observation that people are social, embodied, and embedded in their communities. That’s all dandy, in my view. Of course, his claim is stronger: survival rests on a spectrum. I have a few worries about the latter conclusion.

1. Self-consciousness

First, Parfit fails to adequately address the phenomenology of self-consciousness. Pragmatically, in thought experiments, yes, personal identity is far fuzzier than binary relations. However, there is an element of subjective experience which is immediate and evidently special. Special in the sense that I have privileged access to a set of experiences which cannot not be called “mine.”

G.E.M. Anscombe has an exceptional paper about self-consciousness called “The First Person,” which argues from our use of the concept of ego in language, implied or explicit, infers that our self-conscious, first person experience is ontologically special—human consciousness is in another category from regular objects.³ The world opens up to us collectively, yes, but it also opens up to each person in their experience. This could be the best ground for individual, personal identity, and this criterion seems to imply a black and white picture of what’s important.

This is precisely the view he’s undermining, and the view that a show like Severance questions. But, by raising this objection here, I want to say I’m not fully convinced. Subjectivity and individual experience resist impersonal descriptions. It would take a rigorous philosophy essay to unpack the argument, beyond this essay’s scope (a philosopher’s favorite cop-out).⁴

Worse than these possible metaphysical or ontological misjudgments, his views seem to lead to some pretty horrific ethics.

2. “Repugnant” ethics

Second, some “repugnant” ethical implications of Parfit’s view of identity are spelled out by none other than Parfit himself. This paper only gives a snapshot of his philosophy, and this is the only bit of Parfit I’ve read. According to my friend Cole, a Parfitian nerd, however, Parfit acknowledges that utilitarianism plus his view of identity leads to some admittedly detestable conclusions.

If survival sits on a spectrum, then quality of life seems to be the best criterion of life’s value. So, if we had to choose to save 10,000 people with a high quality of life or 1 million with a low quality of life, we would choose to save the 10,000 people with a high quality of life. That seems… bad—repugnant, according to Parfit.

On the other hand, sometimes our intuition seems to count in favor of his view. For example, when someone who is very old, gets severe dementia and begins losing their character, we say they’re “not themselves,” and in a sense, don’t “survive” as themselves.

On the other other hand, how many lowest-class people, with low quality of life, could an autocrat sacrifice to ethically maintain the high-quality life of his family, friends, and own, in Parfit’s view? Hundreds? Thousands? Tens of thousands?

Hmm. Again, bad.

Parfit’s view of identity seems to miss the ethical weight of the individual, something he himself apparently struggles to accept or explain. And the results can be horrendous.

3. Lack of clarity

Finally, and most interestingly, he chooses psychological connectedness as his base for what grounds identity (making it a spectrum, not binary). Why, though, does he select psychological connectedness? Because it’s “what matters to us.”

Who is the operative “us” here? Western, secularized philosophers, the 0.0000001% of the world’s population? Jokes aside, when we abandon our black and white view of identity, it does seem broadly common sense and integrates well with non-Western worldviews, so I would rather not push this too much.

This vagueness of “what matters to us” and what psychological connectedness actually is, is simultaneously the paper’s weakness and strength—he doesn’t spell out “what matters” to us much, except to say, “what matters isn’t binary identity, like we westerners often think.” Psychological connectedness, I think, it meant to be taken very broadly.

Overall, I think it’s best if we take this paper to be something weaker than a spelled-out version of a view of personal identity. Instead, it problematizes the binary view of identity. I read it as moderately agnostic regarding “what matters,” since he doesn’t spell out what all these connections are.

So, what would Parfit say about Severance?

In thought experiments like Severance, Parfit’s view shines.

Does Mark Scout’s “innie” survive if the innie ceases to wake up? How about his “outie,” if it ceases to wake up? Parfit’s answer: There is no ‘correct” black and white answer, it’s a matter of degree. And this, to me, prima facie, seems correct. That’s why it’s the idea of Severance is so disturbing, because it breaks our categories of personal identity.

“Severance” by Apple TV

Certain psychological connectedness of the innie survives in the outie and vice versa—for example, their personality, speaking habits, tics and quirks, and more, but many, like memories, values, moral character, grief, feelings in general, seem to not transfer. Yet, they’re still continuous with each other, to a degree, just not continuous in their memories (and therefore, certain feelings, beliefs, etc.).

We can then put weight or significance in terms of a spectrum. When an “outie” stops coming in for work, and therefore, their “innie” stops waking up, is it murder? Sort of. Does the innie “die?” No, seemingly. But it’s also not insignificant—the innie survives as the outie only a little bit. As such, it shouldn’t be considered murder, but it’s also probably unethical. We must give the cessation of the innie ethical weight, according to Parfit.

The question of whether they survive? The answer is sort of.

This is precisely, I think, how we feel watching the show. Ending the innie’s memories and existence seems like Murder Lite—not quite fully murder, but not, not killing them either. Perhaps that’s unsatisfying, but I think other answers are even less satisfying.

Parfit’s ideas are fruitful to reflect on, despite my disagreements. I hope it’s been fruitful for you, too.

Is Parfit right? Do you enjoy the show? Does the innie survive, or not at all? (No spoilers in the comments!)

I don’t wholly agree with Parfit about identity, so I’m curious how others, like substance dualists, would answer the Severance problem. Does one soul become two?⁵ Leave your thoughts below, and I’ll respond.

Next essay…

Next essay,⁶ I’m keen to explore another concept. What if, instead of psychological connectedness mattering most to us, there was another criterion, like spirituality, that didn’t map onto physical facts? How might that play out?

I know it’s sounds hippy, a little New Age-y, with a touch of The You You Are vibes, but I’m deadly serious—I think it could be a rigorous, interesting concept.

Stay tuned for more thoughts about Severance and the “spiritual continuum.”

Subscribe for more unorthodox, but orthodox, Christian philosophy

Soli Deo Gloria

1. I’m indebted to my friends, Cole Smith, Hongyu Chen, and Kobe Arbis for their discussions of this paper with me, and for Cole’s recommendation of this essay.

2. Parfit, Derek. “Personal Identity.” The Philosophical Review 80, no. 1 (1971): 3–27. https://doi.org/10.2307/2184309.

3. Anscombe, G. E. M. “The First Person.” In Mind and Language, edited by Samuel D. Guttenplan, 45–65. Oxford University Press, 1975.

4. My argument would probably follow Thomas Nagel, who argues that “perspectiveless,” third-person descriptions of the world cannot account for all the facts in the world, specifically those of consciousness.

5. (Mild spoilers in this footnote!) I’m aware that in one episode, a character says their priest says that innies have new souls. Since he’s been bad in his life, he figured he would get severed so that at least one of his souls would get to heaven (and join his spouse there). A fascinating, and kind of funny, idea.

6. I promised to write another essay about Anscombe, but I’m putting it on hold. I have a rule with myself to only write about what interests me, and this Parfit paper was on my mind. I’m sure I’ll write an introduction to her philosophy soon.