Lecture 3-2: Emotion, Body, and Embodied Personhood
Suppose, for the sake of argument, that a future AI system somehow closed the gap described in the last lecture — suppose researchers became genuinely convinced that its symbol manipulation had become, by whatever criterion satisfies them, real understanding. Would that make it a person in the sense that matters theologically? This lecture argues no, and for a different reason than 3-1's: even granting understanding, a disembodied system still lacks something that is not an add-on to human cognition but woven into how human cognition actually works — a body that can be hurt, that ages, that hungers, and the emotional life that comes bundled with having one.
The argument here has two parts. The first is empirical and philosophical: emotion is not a distraction from good reasoning that a purer, more computational intelligence would do without; it is a necessary component of practical reasoning, and people who lose access to it through neurological damage do not become more rational — they become unable to decide well at all. The second part is theological: the way God chose to be definitively present to humanity was not information transfer but incarnation — taking on an actual body, capable of weeping, capable of dying — and that choice is not incidental background to Christian anthropology but its center of gravity.
이 15분으로 얻는 것
- Explain why the embodied-cognition thesis rejects the picture of the mind as "portable software" that could in principle run on different hardware.
- Summarize Antonio Damasio's somatic marker hypothesis and the clinical evidence behind it.
- Explain why John 1:14 is treated as this chapter's theological center of gravity.
- Connect 1 Corinthians 6:19 and John 11:35 to the argument that bodily vulnerability is not incidental to personhood.
- Identify what, specifically, a text-generation system cannot do in the relevant sense, even when it produces emotionally appropriate-sounding language.
John 1:14 (World English Bible): "The Word became , and lived among us. We saw his glory, such glory as of the one and only Son of the Father, full of grace and truth."
1 Corinthians 6:19 (WEB): "Or don't you know that your body is a temple of the which is in you, which you have from God? You are not your own,"
John 11:35 (WEB): "Jesus wept."
3.1 The picture this lecture argues against
A great deal of popular and even academic writing about minds, both human and artificial, quietly assumes a picture worth naming explicitly so it can be examined: the mind as a kind of software, and the body as a kind of hardware the software happens to be running on at the moment. On this picture, what makes you you is an informational or computational pattern, and that pattern could, in principle, be transferred to different hardware — a different brain, or eventually a computer — without loss, the way a document can be moved from one hard drive to another without changing a single word. This is a modern, more technical descendant of a much older intuition, sometimes traced to Descartes' sharp division between mind (a thinking, non-extended substance) and body (an extended, non-thinking substance): what you essentially are is the thinking part, and the body is where it happens to reside.
This lecture, along with a growing body of research in cognitive science usually gathered under the label "embodied cognition," argues that picture is false — not merely incomplete, but false in a way that matters for how we should think about the relationship between minds and bodies, machine and human alike. The strongest form of the counter-argument does not deny that a great deal of cognition involves manipulating abstract representations. It denies that this manipulation floats free of the body doing it.
3.2 Damasio and the somatic marker hypothesis
The single most influential piece of evidence for this counter-argument comes from clinical neuroscience, and specifically from the work of Antonio Damasio, described at length in his 1994 book Descartes' Error (a title chosen deliberately to name the very picture this lecture is arguing against). Damasio studied patients with damage to the ventromedial prefrontal cortex, a brain region connecting reasoning centers to the body's emotional signaling systems. These patients presented a striking and counterintuitive profile: their logical, abstract reasoning abilities — the kind measured by standard intelligence tests, syllogistic inference, working through hypothetical scenarios — remained fully intact. What they lost was the capacity to make good real-world decisions. One of Damasio's most cited patients could spend an entire session reasoning, in exhaustive and perfectly logical detail, about which of two appointment dates to choose, weighing consideration after consideration, and never arrive at a decision at all; the same patients routinely made catastrophic financial and al choices despite understanding, when asked afterward, exactly why those choices had been bad ones.
Damasio's explanation is the "somatic marker hypothesis": when a healthy person considers an option, the body generates a rapid, largely unconscious emotional response — a "gut feeling" — that marks some options as attractive and others as aversive, well before slow deliberate reasoning has finished working through the details. This somatic marking is what allows practical decision-making to happen at all in realistic time and under realistic uncertainty; without it, a person is left with an unbounded space of logically available considerations and no felt basis for weighting any of them over the others. The upshot, argued carefully and not merely asserted, is that emotion is not an impediment to rational decision-making that a purer intelligence would do better without — it is a necessary component of practical reason as it actually operates in an embodied creature. A system with faultless formal logic and no somatic marking is not a better reasoner than a human; on Damasio's evidence, it is a worse one, specifically at the kind of reasoning that has to end in an actual decision.
This matters directly for how AI should be assessed. A language model has no body to generate a somatic marker, no stake in its own outcomes, nothing at risk when it produces one answer rather than another. Whatever it is doing when it outputs a decision-shaped piece of text, it is not doing the thing Damasio's patients lost and ordinary embodied reasoners retain.
3.3 The Word became flesh: embodiment as the theological center
If chapter 3's first lecture argued that biblical "knowing" resists a purely informational account, this lecture's theological center makes an even sharper version of the same point about God's own self-disclosure. John 1:14 states that "the Word became flesh, and lived among us." The Greek behind "lived" (skēnoō) carries the sense of tabernacling — pitching a tent, taking up bodily residence in a particular place among a particular people — deliberately recalling the tabernacle in the wilderness, where God's presence dwelt in the midst of Israel. The claim is remarkable and easy to read past from familiarity: when God chose the definitive mode of self- to humanity, the choice was not a perfected transmission of information about God. It was a body. Flesh, in the Johannine sense, includes vulnerability, hunger, fatigue, and eventually death — the full weight of what it is to be a bodily creature exposed to a bodily world.
This is the chapter's theological center of gravity, not a supporting text among several, because it makes a claim no other passage in this chapter states as directly: God's own definitive answer to the question "how should ultimate truth be made present to human beings?" was incarnation, not information. If the most complete self-disclosure of God available to humanity took the form of a vulnerable body rather than a body of texts or facts, that is a strong theological signal about what genuine presence and genuine personhood require — and a strong reason to be skeptical of any account, Christian or otherwise, that treats embodiment as an optional or replaceable feature of what a person is.
3.4 The body as temple, and a God who weeps
Two further texts reinforce this from different angles. 1 Corinthians 6:19 tells the Corinthian church, in the middle of an argument about sexual ethics, "don't you know that your body is a temple of the Holy Spirit which is in you, which you have from God? You are not your own." The force of the verse for this lecture is not primarily ethical but ontological: Paul locates the Spirit's dwelling in the body itself, not in a separable mental or spiritual component that merely happens to be housed there for now. The body is not incidental packaging around the "real" person; it is the site of the Spirit's residence. This is a hard verse for any anthropology, Christian or otherwise, that treats bodies as replaceable substrate.
John 11:35, the shortest verse in most English Bibles, carries disproportionate theological weight here: "Jesus wept." At the tomb of his friend Lazarus, whom he was moments away from raising, Jesus does not merely produce comforting words for Mary and Martha — he weeps, genuinely, in a situation whose outcome he already knew would be reversed. This is bodily, involuntary, emotionally costly grief, exhibited by the incarnate Son of God, and it stands as a direct rebuke to the idea that authentic presence with someone in sorrow could be achieved by well-chosen sympathetic language alone. A chatbot can generate text expressing sorrow with perfect grammatical and emotional appropriateness — "I'm so sorry for your loss" arranged in any number of fluent variations — but it has no body to weep with, no stake in the loss, and nothing at risk in producing the words. The category of weeping-with is bodily and costly in a way that text generation, however well-tuned to sound sympathetic, is not, and this verse is chosen precisely because it shows the incarnate God himself operating in that costly, bodily register rather than a purely verbal one.
3.5 What follows for how AI should be described
None of this requires denying that AI systems can produce text that is emotionally appropriate, comforting, or even genuinely useful in the moment — chapter 12 will take up the pastoral-care question in detail and will not dismiss the real, documented usefulness people find in these tools. What this lecture establishes is narrower and more precise: whatever a language model does when it produces emotionally-toned language, it is not doing what Damasio's patients lost and what John 11:35 depicts, because it has no body at stake and no somatic marker to lose. The word "simulate" is doing real work in a phrase like "simulated empathy" — it names a text that has the surface shape of empathetic communication produced by a process that lacks the embodied basis empathy, in the human and the incarnate-divine cases, actually has.
스스로 확인해 보세요
읽었는지가 아니라 설명할 수 있는지를 묻는 것입니다. 체크는 이 기기에만 저장됩니다.
묵상 및 토론
- Explain Damasio's somatic marker hypothesis in your own words, including what the vmPFC patients could and could not do. Why does this evidence count against, rather than for, the idea that emotion is an obstacle to good reasoning?
- Why does this lecture single out John 1:14, rather than another Christological text, as the "theological center of gravity" for the whole chapter? Do you find that framing persuasive?
- What work is the word "temple" doing in 1 Corinthians 6:19 that a phrase like "the body houses the soul" would not do?
- Jesus, in John 11:35, weeps at a death he is about to reverse. What does the timing of his grief — before the miracle, not instead of it — suggest about why bodily, emotional presence is not made redundant by having the power to fix the underlying problem?
- Can you think of a case where an AI system's emotionally appropriate output was genuinely helpful to someone, and also a case where mistaking that output for real empathy caused harm? What distinguishes the two cases?