Lehrer vs Elster on love

Compare these texts and write a synthesis of what we learn from the comparison.

[The prompt here contained the entirety of “Love” (§IV.3 of “Ulysses and the Sirens”, by Jon Elster) and “Love and Transcendence” (Chapter 5 of “Self-Trust”, by Keith Lehrer), which are omitted from this post for copyright reasons.]

I gave this prompt to both Gemini and ChatGPT.

• Gemini
• ChatGPT

Gemini 3 output

This synthesis compares Jon Elster’s "Ulysses and the Sirens" (§IV.3) and Keith Lehrer’s "Self-Trust" (Chapter 5).

Both authors grapple with a central philosophical dilemma: The tension between Love and Autonomy (Freedom). Both take Jean-Paul Sartre’s pessimistic view—that love is a contradictory attempt to possess another’s freedom—as a starting point. However, they diverge significantly in their conclusions: Elster explores why love is often an "inherently unrealizable project" fraught with paradox, while Lehrer constructs a logical hierarchy to prove that "Autonomous Love" is possible through self-regulation.

---

1. The Shared Diagnosis: The Paradox of Possession

Both authors agree that love, in its raw or "extreme" form, creates a logical contradiction regarding freedom.

• The Sartrean Problem: The lover wants to possess the beloved. However, they do not want to possess a robot or a slave; they want to possess the beloved's free choice to love them.
• The Contradiction: To demand someone love you is to destroy their freedom to choose. As Elster notes via the "Be Spontaneous" paradox: "I want you to love me freely, but I demand that you do so."

2. Jon Elster: Love as an Unstable Game

Elster treats love primarily as a psychological knot and a game-theoretic problem. He focuses on why love creates instability, decay, and "clinical symptoms" of irrationality.

• The Seesaw of Reciprocity: Elster highlights the "Groucho Marx paradox": I wouldn't love anyone who would stoop so low as to love me.

• If the beloved does not love back, the lover is obsessed.

• If the beloved does love back, the lover loses interest because the beloved has "lost value" by engaging with an inferior.

• Game Theory & Dominant Strategies: Elster argues that love is like a game without a "dominant strategy" (a move that is always best). For love to work, both parties must transform simultaneously. However, this is precarious.

• The Collapse of Time: Elster introduces the element of time as a destroyer of love.

• Anticipation of Decay: Like a Prisoner’s Dilemma, if lovers know love might end (decay), they may preemptively withdraw to protect themselves, causing the very decay they feared.

• Inner Time: Stendhal’s observation that the waiting lover consumes the present moment with past expectations and future fears, making the actual meeting with the beloved disappointing.

Key Takeaway from Elster: Love is often an "inherently unrealizable project" because the dynamics of desire, time, and strategic interaction tend toward instability and paradox rather than equilibrium.

3. Keith Lehrer: Love as a Hierarchy of Preference

Lehrer accepts the danger of "bondage" in love but rejects the idea that it is inevitable. He proposes a taxonomy of love, moving from base desire to high-level autonomy. His solution is Metamental Ascent: the ability to think about and evaluate one’s own thoughts.

Lehrer distinguishes between Desire (first-order, impulsive, bondage) and Preference (higher-order, evaluated, autonomous).

Lehrer’s Taxonomy of Love

1. Extreme Love (Bondage): Based on raw desire. "I desire to satisfy your desires." This creates a feedback loop (sniffing desire in roses) but results in slavery to the other’s whims.
2. Desire/Preference Love (Asymmetrical): One party loves based on desire, the other on preference (e.g., parent/child).
3. Fully Autonomous Love (The Solution): This creates a "keystone loop" of stability.
• Definition: S prefers A out of fully autonomous love for T if and only if S autonomously prefers A because T autonomously prefers A.
• The "Trick": How do you avoid bondage? You prefer what the partner prefers because you prefer to have that preference structure. You are not obeying them; you are obeying your own decision to align with them.

4. Synthesis: Pessimism vs. Constructivism

The comparison reveals a fundamental divide between a descriptive analysis of human failure (Elster) and a normative framework for human success (Lehrer).

Feature	Jon Elster ("Ulysses and the Sirens")	Keith Lehrer ("Self-Trust")
View of Love	A paradox; an "unrealizable project."	A manageable structure; a "transcendence."
Role of Time	Destructive. Future decay ruins the present.	Constructive. Trust implies stability over time.
Mechanism	Game Theory: Prisoners Dilemma, Dominant Strategies.	Preference Logic: S prefers A because T prefers A.
Autonomy	Threatened by love. The lover tries to be both subject and object (en-soi and pour-soi).	Secured by love (if autonomous). We "open the loop" of self-trust to include another.
Outcome	Likely instability, "seesaw" dynamics, or decay.	Potential for "Autonomous Love" based on mutual trust and reason.

Conclusion

From comparing these texts, we learn that desire is the source of the paradox, but preference is the source of the solution.

Elster remains stuck in the "desire" phase (which he calls "sentimental stuff" or clinical symptoms), where the clash of freedoms leads to a "game without a solution." Lehrer argues that we can escape this game by moving up a level: by converting raw desires (which push us around) into preferences (which we choose to hold).

Synthesis: Love is a paradox of bondage only if we remain passive subjects of our immediate desires (Elster’s view). If we exercise "metamental" control—choosing to value the other's autonomy through our own autonomous choice—love becomes a stable expression of self-trust (Lehrer’s view).

---

Next Step: Would you like me to construct a formal truth table or logic diagram applying Lehrer's definition of "Autonomous Love" to one of Elster’s "Seesaw" scenarios to see if it solves the paradox?

ChatGPT 5.2 output

Reading Elster beside Lehrer teaches you to see “love” as less a simple feeling than a dynamic structure—a set of feedback loops between two agents—where the same mechanisms that generate intensity also generate instability.

What both texts agree on

Both authors treat love as essentially relational and reflexive: what matters is not only what I feel, but how my feelings/choices depend on your feelings/choices, and how that dependence feeds back.

• Reciprocity is central but dangerous. Each treats reciprocation as the “object” of love while also diagnosing ways it can undermine love. Elster’s “seesaw” (wanting recognition that loses value when granted) mirrors Lehrer’s “magic turns to bondage” (mutual desire becomes dependence, then manipulation and resentment).
• Love tangles with freedom. Both lean on Sartre’s idea that love involves wanting the other’s freedom as freedom and yet wanting it to yield to us. Elster presents this as a paradox (“possess a liberty”); Lehrer reframes it as a conflict between love and autonomy that can be redesigned rather than simply endured.

Where they sharply diverge

1) Elster is diagnostic; Lehrer is reconstructive

• Elster is mainly explaining why love becomes paradoxical: he inventories structural contradictions (Groucho Marx paradox, “be spontaneous,” game-theoretic deadlocks, anticipation reversing time, Newcomb-like self-scrutiny). Love is often an “inherently unstable project,” and his emphasis is on failure modes.
• Lehrer wants a solution concept: he builds a taxonomy from desire-love up to fully autonomous love, defined in terms of higher-order preferences (preferring what the other autonomously prefers, because you prefer to have that preference). He’s not denying pathologies—he’s offering a normative architecture meant to avoid them.

2) They disagree on what stabilizes love

• For Elster, a stable love relationship is conceivable when lovers are mutually and conditionally transformed—each changes the “boundary conditions” of the other’s life in a way that enables self-redefinition. The stability is precarious (like certain strategic equilibria) and constantly threatened by the value-destroying effects of reciprocation, time, and self-consciousness.
• For Lehrer, stability comes from autonomy and trust: replacing first-order desire dependence with higher-order evaluation and self-trust extended to the other. The key move is metamental: you don’t merely chase the other’s desires; you endorse (or don’t) the pattern of endorsement itself.

3) Elster foregrounds time; Lehrer foregrounds levels

A distinctive contribution of Elster’s chapter is that love’s paradoxes are not only interpersonal but temporal:

• love can require growth to exist,
• anticipation of decay can cause instantaneous collapse,
• inner time can “collapse” so the present is experienced as a junction of past expectation and future regret.

Lehrer’s distinctive lens is hierarchy of attitudes:

• desire → preference → autonomous preference,
• and the way each level changes what “bondage” looks like (bondage to desire vs. “higher-order bondage” to another’s preferences).

The synthesis: love as a two-person self-referential system

Put together, the comparison yields a powerful composite picture:

1. Love generates feedback loops that are intrinsically volatile. Elster maps the classic loop: I want your recognition because it has worth only if not guaranteed; I want your spontaneity but also want it directed at me; I want certainty about what will avoid regret, but self-monitoring corrupts the very spontaneity/candor love needs.

2. Trying to “solve” love at the same level that produces it often worsens it. Elster’s examples show that direct control strategies (“I’ll act so as to get you to…”) easily backfire by converting love into a game of inference, signaling, and second-guessing—where sincerity becomes unreadable and the present gets colonized by future-oriented anxiety.

3. A plausible escape is a shift in what is being coordinated. Here Lehrer supplies a missing design principle: stabilize love not by controlling the other’s desire, but by coordinating autonomous preferences—with trust doing the work that control cannot. This doesn’t remove vulnerability (Lehrer admits autonomous love can feel insecure precisely because it’s not capturable), but it changes the kind of vulnerability: from bondage/manipulation to risk under trust.

4. Yet Lehrer’s “solution” also inherits Elster’s warning. Elster helps you see why even “autonomous love” can be fragile: higher-order endorsement can itself become performative, strategic, or time-haunted (“Will I regret endorsing this endorsement?”). Lehrer anticipates this in his discussion of “love knots” (wanting to be loved non-autonomously) and in the instability of giving the other’s preferences too much weight (cycling, self-erasure).

What we learn overall

Elster teaches you why love so often collapses into paradox: it is a project aimed at another’s freedom, recognition, and spontaneity—goods that lose value when produced by force or guaranteed by design, and that are distorted by anticipation and self-scrutiny. Lehrer teaches you how love might be made livable: by moving from desire-driven dependence to preference-driven evaluation, then to mutual autonomy and trust, while recognizing that even this “highest” form has its own tensions (vulnerability, scope limits, weight-adjustment problems).

If Elster is the anatomy of love’s contradictions, Lehrer is an attempt at love’s ethics and engineering—and the comparison suggests that any “engineering” that forgets Elster’s paradoxes will accidentally rebuild bondage at a higher level.