I expected Éric Rohmer’s Conte d’hiver (A Tale of Winter, 1992) to leave me thinking about romantic fidelity, chance, and perhaps a little theology. Instead, I ended up arguing with Félicie, defending her, criticizing her again, correcting some of my own criticisms after returning to the screenplay, questioning whether her “faith” is really faith, and finally becoming much more interested in Loïc than in the miraculous return of Charles. The film grew richer every time one of my first explanations stopped working.
That is why I do not want to write this as though I watched the film once and immediately possessed a finished interpretation. I did not. My reading developed by repeatedly testing claims against chronology, dialogue, action and consequence. At first I thought Félicie had spent five years simply waiting. The screenplay forced me to withdraw that claim. I initially thought one conversation with Maxence looked like straightforward hypocrisy; later I found motivated self-deception a more precise category. At first the central theological question seemed to be whether Félicie’s conviction deserved to be called faith. By the end, another question had become at least as important: if one person’s hope is fulfilled, what happens to the people who carried the costs of that hope?
This is also why Loïc became the character who affected me most. Charles is the lost beloved who returns. Félicie is the believer whose astonishing hope is narratively vindicated. Loïc is the person who remains present, helps, listens, suffers, prays, and then largely disappears from the story once the miracle arrives. The film says: the lost person has been found. I increasingly wanted to ask: yes, but who has now been left outside the frame?
I treated the film as an argument that had to survive its own evidence
Rohmer’s films invite this kind of reading because conversation does not merely communicate plot. His characters reason, rationalize, contradict themselves, correct themselves and often build elaborate explanations around feelings they only partly understand. Imogen Sara Smith describes Rohmer’s cinema as deeply interested in rationalization, self-deception and the “mystery of subjectivity,” while also emphasizing how difficult his characters find it to understand themselves and one another (Smith, 2024). That description fits A Tale of Winter almost perfectly.
I therefore ended up using something resembling a debugging procedure. That may sound slightly absurd for a Rohmer romance, although frankly the characters themselves spend enough time debugging their own emotional state that I do not feel entirely responsible for the metaphor.
claim → exact dialogue or visual evidence → chronology → competing interpretation → action produced by the belief → consequences for other people → philosophical or theological test → what remains unproven
The chronology mattered especially. The cathedral experience comes before the performance of Shakespeare. Félicie’s attempt to commit herself to Maxence comes before the cathedral. Her practical attempts to recover Charles occurred years earlier. Pascal and Plato arrive later as conceptual vocabulary supplied largely by Loïc. If I flatten those events into “Félicie believes in Charles and is rewarded,” most of the film’s intellectual machinery disappears.
The method also forced me to distinguish several kinds of evidence. Dialogue establishes what a character says, but not necessarily what is true. Behaviour can confirm or contradict speech. Visual staging can complicate both. Later outcomes tell us what happened, but they do not necessarily prove that an earlier belief was rationally justified. External philosophy and theology can illuminate the categories the film invokes, but they should not be used to overwrite what the film itself actually shows.
This is particularly important because some influential descriptions of the film already use the word “faith.” Criterion describes Félicie as holding an “undying faith” that a miracle will reunite her with Charles, while La Cinémathèque française speaks of her certainty and wager on love gradually becoming faith (Criterion Collection, n.d.) (Jamet, 2019). Those are illuminating descriptions. I nevertheless think the word becomes much more interesting once we ask what kind of faith this actually is.
The first crack appears in a verb
One of the scenes that changed my attitude toward Félicie occurs before any explicit theological discussion. She is with Maxence, trying to persuade herself that Charles belongs to the past. She says:
« Je ne veux plus penser qu’à toi, je ne veux aimer que toi. »
Maxence hears the problem immediately:
« Tu “veux” m’aimer ? »
Félicie answers that she does love him, but would like to love him more (Rohmer, 1992).
The grammar is revealing. She does not simply state an emotion. She describes an act of will directed toward an emotion. She wants herself to love Maxence exclusively. The screenplay then has her move physically into his arms. Her body performs an intimacy that her sentence has already qualified.
My first reaction was severe: is this not simply hypocrisy? She knows Charles remains central, yet tells Maxence that she wants to think only of him and love only him. After sitting with the scene, I became less certain. The concept of motivated self-deception seems more accurate. She may not be calmly lying to Maxence while privately knowing the opposite. She seems to want a proposition about her future to become true and speaks partly from inside that desired future.
This is more charitable than accusing her of calculated deception, but it does not make the problem morally harmless. Self-deception can injure another person just as effectively as deliberate deception. Sincerity in a particular moment is not a complete defence when somebody else is being asked to organize his life around the statement.
The scene also became important because Maxence himself detects the discrepancy. My interpretation does not require an external psychological theory imposed on the character. Rohmer puts the objection directly into Maxence’s mouth: you want to love me?
That tiny grammatical distinction becomes a prototype for the whole film. Again and again I found myself separating:
- what Félicie wants to be true;
- what Félicie genuinely feels;
- what Félicie believes she knows;
- and what is actually true outside her consciousness.
The film eventually makes some of those categories converge. My question is whether the convergence at the end should be allowed to erase the distinctions that existed before it.
I initially accused Félicie of passive waiting and the screenplay proved me wrong
This was one of the most useful corrections in my reading. I initially found Félicie’s hope frustrating because I thought: if Charles is so important that she is willing to organize five years of her life around his possible return, why does she not actually search for him?
That criticism was too simple.
The screenplay establishes that she did search. The original disaster begins because she tells Charles that she lives at 36 rue Victor-Hugo in Courbevoie when she actually means Levallois. She does not discover the mistake immediately; she realizes it months later while dealing with maternity paperwork. Once she understands what happened, she goes to the post office in Courbevoie, asks what might have happened to letters, checks poste restante, discovers that Courbevoie also has a rue Victor-Hugo, and investigates the corresponding address. Later, after meeting Loïc, she says that he helped by writing to many organizations (Rohmer, 1992).
Then comes the almost absurd informational constraint: she does not know Charles’s surname. He had been working informally at the restaurant where they met, everybody called him Charles, and the surname—possibly Danish or Dutch—never remained in her memory. Searching across continents with no surname, uncertain geography and no stable institutional record is very different from simply refusing to open a phone book.
I therefore had to withdraw the strong claim that Félicie simply “does nothing.” The evidence does not support it.
But withdrawing that criticism did not end the problem. It changed it.
Once the available search mechanisms have largely collapsed, what exactly does her hope do? The answer seems to be that she begins preserving a different kind of possibility. She cannot meaningfully increase the probability of finding Charles, but she can avoid entering commitments that would make it harder to receive Charles if he reappeared.
In decision-theory language, this resembles the preservation of an option:
Early problem: maximize P(find Charles) Later problem: preserve P(being free to receive Charles | Charles returns)
That distinction helped me understand the cathedral decision. Félicie does not emerge from the church with a new search algorithm. She emerges with a new commitment policy.
After returning to Maxence, she says that she has decided she will no longer bind herself to men except provisionally. She is trying to prevent another relationship from closing the branch of the future in which Charles returns.
This is much more coherent than my original description of simple passivity.
Yet it leads directly to another question: who pays for the preservation of that option?
The cathedral gives her clarity but does it give her knowledge?
The cathedral scene became one of the most important parts of the film for me, partly because it is frequently interpreted too quickly as a religious conversion. Félicie herself resists that description.
She has been upset by Maxence, leaves with Élise, passes the cathedral in Nevers and sits down while the child looks at the crèche. When Loïc later asks whether she prayed, she corrects him. It was not prayer in the conventional sense. It was closer to reflection or meditation. She describes the familiar state in which lack of sleep, decision pressure and mental excitement can make thoughts run unusually quickly, except that this experience was “a thousand times” stronger. Suddenly everything became clear (Rohmer, 1992).
Then comes the sentence I kept returning to:
« J’ai pas pensé : j’ai vu, j’ai vu ma pensée. »
She says she “saw” what she had to do and “saw” that she was not mistaken.
I find the phenomenology fascinating. I also find the epistemology dangerous if the two are confused.
At least three propositions are involved:
- Félicie genuinely experiences extraordinary subjective clarity.
- The conclusions experienced in that state correspond to external reality.
- The experience is properly described as theological faith or divine revelation.
The first seems strongly supported. Nothing in the film requires us to think she is faking the experience.
The second does not follow automatically. “I saw my thought” is an unusually vivid report about consciousness. It does not provide data about Charles’s current location, relationship status, health, desires or future movements.
The third is even less straightforward, because Félicie explicitly tells Loïc that she was not thinking of God. When Loïc romanticizes the experience and says that if he were God he would cherish her for sacrificing happiness to an absent love, Félicie replies that if God loves her, God should give Charles back. But she immediately adds that she had not actually been asking God for anything. During the cathedral moment, she had felt “alone in the world, alone in the universe,” and concluded that it was up to her to act (Rohmer, 1992).
This is why I eventually preferred the phrase existential illumination in a religious space. It preserves the seriousness of what happened without pretending that the film has established a supernatural causal mechanism.
And something in the insight may indeed be true. Félicie realizes that she should not permanently bind herself to somebody she does not love “à la folie” merely because she is frightened by uncertainty. The strongest content of her revelation may therefore be negative:
I should not make an irreversible commitment to a life I do not actually want simply because I am afraid that no better future will come.
That is a much stronger proposition than “Charles will return.”
One concerns the truth of her own present commitment. She has considerable evidence about that.
The other concerns an absent man’s future return. She has almost none.
Subjective certainty is not the same thing as external truth
This distinction eventually became the epistemological centre of my reading. Félicie repeatedly treats the force of an inward experience as though it reduced ambiguity in the external world. Her language is visual: she did not merely think; she saw. She did not merely choose; she saw that she was not mistaken.
In computational terms, I would separate confidence from calibration. A model can output 0.99 confidence and still be wrong. Human beings discovered overconfidence long before machine learning; neural networks merely automated one of our older traditions.
The same problem appears in moral and religious experience. Catholic teaching on conscience does not say that the strongest inward conviction automatically constitutes truth. The Catechism of the Catholic Church defines conscience as a judgment of reason concerning the moral quality of a concrete act and explicitly acknowledges that conscience can make erroneous judgments. It also recommends interpreting experience with prudence, competent advice and serious discernment (Catholic Church, 1992, §§1777–1788).
This makes the cathedral scene more interesting, not less religious. A serious theology does not need to dismiss interior experience. It asks what kind of claim the experience can support.
I therefore increasingly wanted to say:
The intensity of certainty is evidence about the subject who experiences it. It is not automatically evidence proportional to the certainty of the proposition believed.
This is precisely where I begin resisting attempts to canonize Félicie’s certainty simply because the screenwriter later makes Charles appear.
Shakespeare recognizes the structure she already carries
The chronology matters again. Shakespeare does not create Félicie’s cathedral experience. Loïc takes her to The Winter’s Tale afterward. In Shakespeare’s final scene, Hermione—believed dead for sixteen years—appears as a statue. Perdita kneels before her. Paulina prepares the spectators for something astonishing, warns against interpreting it as wicked magic, and eventually tells them that they must “awake” their faith. Music sounds; Hermione moves and descends (Shakespeare, n.d./2016, 5.3).
Félicie is overwhelmed. She grips Loïc’s hand and later says she almost cried out when the statue moved. Loïc’s response is revealing because he remains interested in the text’s ambiguity. Was this literally supernatural animation, or had Hermione somehow never died? Félicie responds:
« Je n’aime pas ce qui est vraisemblable. »
Then she insists that the meaning is obvious: faith makes Hermione live again (Rohmer, 1992).
I initially found the stage miracle less affecting than Félicie does. As a spectator, I know that the “statue” is an actor who was standing still and then moves. But that objection turned out to be too literal. Félicie is responding to the narrative structure, not to special-effects technology. Someone is lost. Years pass. The loss seems irreversible. The impossible return occurs. That structure maps almost perfectly onto Charles.
Shakespeare therefore functions as a révélateur. It gives an aesthetic image to a pattern Félicie already carries. Scholarship on Rohmer has likewise treated Shakespeare as structurally central to the film’s movement from winter, loss and apparent death toward restoration (Smith, 2024). But I remain much closer to Loïc than to Félicie in the argument afterward. He says there is ambiguity. She says there is none. That difference is epistemologically important.
Félicie thinks her interpretation is religiously superior because it requires less hesitation. I am not convinced. Religious confidence and theological maturity are not identical. Sometimes the more theological response is precisely the willingness to say:
“I am moved by this, but I still do not know exactly what happened.”
Her anti-intellectualism is attractive and also dangerous
The tension between Félicie and Loïc is partly social and intellectual. Loïc reads. He reaches instinctively for Pascal, Plato, Shakespeare and philosophical explanation. Félicie experiences this as mediation. At one point she tells him, with wonderful irritation, that if she said she loved him, he would probably go and check whether the statement had already been written in Shakespeare. For him, she complains, only what is written seems true (Rohmer, 1992). The criticism is funny because it contains some truth. Loïc can intellectualize life. Books can become a defensive layer between experience and vulnerability.
But Félicie’s alternative can become equally dogmatic. “I feel directly” is not inherently more truthful than “I have reflected.” Anti-intellectual authenticity can simply relocate the authority from texts to affect. Her strongest formulation is that in the cathedral she felt completely herself, just as she had felt completely herself five years earlier with Charles. That is psychologically meaningful. It tells us which experiences she integrates most strongly into her sense of identity. Yet identity is not infallibility. “This feels most like me” cannot by itself establish “therefore my interpretation of the world is correct.”
That distinction matters well beyond Rohmer. There is a contemporary temptation to treat authenticity as moral certification: if an action expresses my deepest self, questioning it becomes almost an attack on my existence. I think Rohmer is subtler than that. Félicie’s authenticity is powerful, but the film also gives us enough evidence to ask what that authenticity does to other people.
Pascal arrives after Félicie has already made the wager
One of the film’s great pleasures is that Félicie frequently has an experience first and receives a philosophical label later. Loïc is effectively running a conceptual package manager for her. Plato? Installed. Pascal? Installed. Shakespeare? Already in production.
Félicie admits that the chances of recovering Charles are extremely small. She knows he may be married. She knows he may no longer love her. Yet she says that if she did recover him, the joy would be so enormous that she could give her life for that possibility. Then comes the line:
« Vivre avec l’espoir, je pense que c’est une vie qui en vaut bien d’autres. »
Loïc immediately recognizes a Pascalian structure (Rohmer, 1992).
Rohmer himself repeatedly returned to Pascal’s wager, and scholarship on the film has described Félicie’s position as a distinctly Pascalian wager transposed into romantic life. An OpenEdition study makes a particularly useful distinction: Félicie’s belief is no longer straightforwardly Christian—she herself says she was not thinking about God—yet it retains a Pascalian logic of risk, uncertainty and an immense hoped-for good (Amiel, 2013). This helped me clarify something. Félicie’s wager is not Pascal’s theological wager in any strict sense. Charles has taken the structural position occupied by infinite gain, but Charles is still a finite human being. That substitution is philosophically fascinating and theologically unstable.
Her hope is rational in one sense and badly calibrated in another
I do not think Félicie’s position should simply be dismissed as irrational. Once she recognizes that she does not genuinely want a permanent life with Maxence, refusing permanence may be perfectly rational. Choosing security simply because the desired future is improbable can create a different kind of dishonesty. The strongest charitable reading of her position is therefore:
I would rather accept uncertainty than make a permanent commitment I know is emotionally false.
I think that is serious.
The weaker version is:
Because my hope feels absolute, its fulfilment possesses some privileged likelihood or destiny.
Here I become much more skeptical. The film is interesting because Félicie moves between these positions without always distinguishing them. Sometimes she defends freedom from a commitment she does not authentically want. At other moments she speaks as though her intuition gives her special access to the future. Those are different arguments. I can defend the first while rejecting the second.
Fidelity is not as simple as monogamy
Another difficulty I had to resolve concerned whether Félicie can meaningfully be called faithful to Charles while sleeping with and loving other men. The answer depends on what “fidelity” means. If fidelity means sexual and relational exclusivity, then obviously no. Her life contradicts that definition. But the film constructs another kind of fidelity: ultimate affective orientation. Loïc and Maxence occupy real places in her life, yet neither occupies the place Charles occupies. The deepest allegiance of desire never transfers.
This explains how Félicie can sincerely say that she loves Maxence and still mean that Charles remains incomparable. She does not necessarily use the word love univocally. I do not need to endorse that hierarchy in order to recognize its internal consistency. The moral question therefore shifts away from “Is she monogamous?” toward “Has every person involved understood what kind of relationship he is actually in?”
This is where freedom and responsibility intersect. A non-exclusive relationship can be perfectly ethical if the relevant expectations and boundaries are genuinely shared. The difficulty in Rohmer’s film is that the emotional architecture is not symmetrical. Charles has an absolute status that Loïc and Maxence cannot meaningfully compete with, yet both men at different times are invited into forms of intimacy that can feel more definitive than Félicie ultimately allows.
Charles has one enormous advantage over Loïc and Maxence — he is absent
The more I thought about Charles, the less simple Félicie’s “absolute love” became. Charles is absent for five years. Absence is an astonishingly effective way to avoid disappointing someone. Loïc and Maxence must exist as actual human beings. They have habits, temperaments, jealousies, intellectual quirks, professional lives, needs and bodies. They can say the wrong thing at breakfast. They can become irritated. They can fail to match an ideal at 4:17 on an ordinary Tuesday. Charles cannot. He exists for most of the film as memory.
Rohmer reinforces this visually. The opening romance is summer: physical intimacy, brightness, seaside pleasure, an almost embarrassingly concentrated idyll. Five years later, the main film takes place in grey winter streets, trains, suburban houses and salons. Criterion explicitly notes the contrast between the idyllic summer prologue and damp December, while La Cinémathèque describes winter’s greyness as central to the film’s emotional texture (Smith, 2024) (Jamet, 2019). Félicie therefore compares imperfect present men with an absent beloved preserved inside summer memory.
That does not make Charles unreal. He existed. The love existed. Élise is the embodied consequence of that love. But the Charles who structures five years of hope is partly a remembered object protected from ordinary temporal erosion. This made me wonder whether Félicie’s absolute fidelity is partly fidelity to Charles and partly fidelity to the idealized lost possibility that Charles represents.
The ending conveniently removes the problem by returning a Charles who is almost maximally compatible with the memory. He has not forgotten her. He is available. The woman beside him is not his wife. He does not arrive with another child and a complicated family system. He accepts Élise. He still loves Félicie. Rohmer could have returned Charles married, indifferent or fundamentally changed. He does not. The fairy tale gives Félicie the version of reality in which the preserved ideal survives contact with the present.
I admire Félicie’s freedom more than my criticism may initially suggest
My criticism of Félicie should not be mistaken for a conservative objection to her sexual autonomy. In fact one aspect of the film I find attractive is how much freedom its relationships contain. Loïc is hurt when Félicie chooses Maxence. He becomes dry, withdraws and wants the separation to be clean. He is not an emotionless saint. Maxence becomes angry when Félicie leaves him and at one point physically grabs her. Félicie immediately demands release, says he is hurting her, and threatens to hit him if he does not let go. When he releases her, she explains that she physically cannot tolerate being held by force (Rohmer, 1992).
That scene matters because the men should not be idealized into infinitely patient moral furniture. Maxence can become controlling. Loïc experiences jealousy and pain. Félicie’s insistence on her freedom therefore responds to genuine pressures, not imaginary ones. I also think her cathedral insight contains something ethically valuable: she does not owe Maxence permanent commitment simply because he offers a plausible life. Security cannot manufacture eros.
Where I become critical is at another point:
Freedom does not cancel relational responsibility.
It is possible to defend Félicie’s freedom completely and still ask whether she adequately recognizes the costs borne by people who keep accommodating that freedom.
Who subsidizes her authenticity?
This was the question that radically changed my moral reading:
Who subsidizes her authenticity?
Félicie’s freedom is highly visible. She can move between Paris and Nevers, reject Maxence, return toward Loïc, preserve Charles as her deepest horizon, seek emotional support and remain open to an improbable future. But freedom is not exercised in a vacuum. Other people absorb consequences.
Criterion’s Imogen Sara Smith makes a strikingly severe judgment here, saying that Félicie treats both men badly: she abandons Maxence shortly after moving in with him, leans on Loïc after rejecting him, and finally asks Loïc to pray for the return of Charles (Smith, 2024). I would qualify that judgment because the film gives Félicie more moral awareness than the phrase “treats them horribly” might suggest. She knows she is hurting Maxence. She explicitly feels ashamed about putting him in an embarrassing position. With Loïc, she sometimes tries to persuade him that her departure might free him to find someone who can love him differently. She is not completely blind to others.
Yet the centre of gravity remains intensely first-personal. Her moral vocabulary repeatedly returns to:
- what I feel;
- what I saw;
- what I cannot accept;
- whom I love;
- whom I cannot love enough;
- what future I must keep possible.
The much rarer sustained question is:
What does maintaining my authenticity require other people to carry?
This distinction became important to me because authenticity can acquire an almost sacred status in modern moral language. “I must be myself” sounds self-validating. But one person’s authenticity can depend on another person’s emotional labour, forgiveness, availability or patience. A local rule can be perfectly understandable while the global system becomes pathological. Félicie preserves her freedom. Loïc remains available. Maxence tries to accommodate. Each local action has an explanation. The aggregate cost is much harder to see. Apparently distributed systems and French romance share more architecture than expected.
Loïc is not infinitely available and the film gives us evidence of his pain
One thing I wanted to correct in my own admiration of Loïc was the temptation to turn him into a purely selfless saint. He is not. When Félicie first chooses Maxence, Loïc is plainly hurt. He wants the separation to happen quickly because prolonging it has allowed him to maintain illusions. Félicie tells him she will miss him deeply as a friend and even argues that by leaving she may allow him to find “the woman of his life.” Loïc responds that he does believe she means this sincerely. He even explains that part of why he loves her is that he feels he can read her heart (Rohmer, 1992).
Then comes an extraordinary little exchange. Félicie admits that she sometimes lies. Loïc answers, essentially, that she does not lie about the important things. She does not want him to idealize her. He cannot quite stop doing so. So Loïc has his own interpretive vulnerability. His belief in Félicie’s transparency may itself be partly romantic idealization.
That matters because ethical analysis should not distribute all complexity to Félicie and turn everyone around her into perfectly calibrated observers. Loïc loves. Therefore Loïc also interprets. And sometimes love interprets generously beyond what the evidence requires.
The prayer scene became the moral centre of the film for me
The scene that affected me most is astonishingly quiet. Loïc and Félicie are near a church. She tells him that since he is Catholic, he should go to Mass. He says he might if he were alone but does not want to impose it on her. She insists that he should not sacrifice his convictions for her. Then she says that he knows how to pray, so he should pray for her.
Loïc initially says he will pray another day. Félicie refuses:
« Aujourd’hui. Et tu prieras pour moi. »
Loïc answers:
« Je prie toujours pour toi. »
She pushes further. This time she wants him to pray in her place, asking for what she would ask even if he disagrees. Then she admits:
« Tu vois je suis très exigeante. »
Loïc eventually says that in a sense this is already what he does: he prays for her happiness even when that happiness is not necessarily his own (Rohmer, 1992).
This is where I stopped thinking mainly about the beauty of Félicie’s faith and started thinking about the cost of Loïc’s love. What exactly is she asking? In practical emotional terms, something like:
Love me enough to pray sincerely for the return of the man whose return will eliminate your remaining possibility of being with me.
That is enormous. Félicie knows it: “Je suis très exigeante.” Loïc knows it too. He tells her that she is asking a great deal. The beauty of the scene therefore belongs primarily to the costliness of Loïc’s response. Félicie’s request is ethically more ambiguous.
Extraordinary love does not create a right to romantic reciprocity
At this point a dangerous argument could easily appear: Loïc loves so well that Félicie should love him back. I do not think that follows. Love is not distributive justice. Romantic or erotic desire cannot be awarded as moral compensation. Goodness does not generate a claim on another person’s body, sexuality or deepest affection. If Loïc gives ten units of self-sacrifice, Félicie does not owe ten units of eros. That would make love transactional in precisely the wrong way.
Yet the opposite extreme is equally unsatisfactory. Because care creates no entitlement to erotic love, it does not follow that care creates no moral claims at all. There can still be obligations of recognition, gratitude, truthfulness, proportion and responsible boundaries. The real issue is less:
“Will you reward me for loving you?”
and more:
“Do you actually see me, including what your reliance on my love costs?”
That distinction became decisive for me.
Eros, agape and why Loïc must also be allowed to receive
Benedict XVI’s Deus caritas est unexpectedly became one of the most useful theological texts for understanding Loïc. The encyclical refuses a crude opposition between eros and agape. Eros is transformed as it increasingly seeks the good of the beloved, while agape prevents love from collapsing into pure possession. Yet Benedict also makes a point that is crucial here: a human being cannot live by self-giving, descending love alone. The person who gives must also receive love as gift (Benedict XVI, 2005, §7).
That is exactly why I resist a Christianized interpretation in which Loïc’s role is simply to keep sacrificing forever. If we say:
“Because Christian love is selfless, Loïc should accept whatever emotional role Félicie needs and expect nothing.”
we have converted charity into self-erasure.
A healthier formulation is:
Agape without justice can become exploitation.
Eros without reciprocity can become misery.
Self-sacrifice without discernment can become self-erasure.
Loïc is allowed to need something too. He is allowed to say that remaining a lover, fallback, philosophical guide, trusted confidant and emotional refuge while Félicie’s deepest fidelity remains elsewhere is too painful. Christian love does not require becoming permanently consumable. Loïc sometimes behaves as though he is operating a twenty-four-hour emotional support service with no maintenance window. Even excellent infrastructure eventually requires downtime.
The boundary Loïc perhaps should draw
My admiration for Loïc therefore changed during the discussion. At first I admired the fact that he stays. Eventually I wanted a more mature version of staying. He could truthfully say something like:
“I want your happiness. I do not want to control you. But I cannot remain indefinitely in a role where I provide the intimacy of a partner while accepting that your deepest commitment is elsewhere.”
That would not contradict love. It might protect love from becoming distorted.
This led me to a formulation that I now prefer to simple praise of his self-sacrifice:
Loïc’s generosity without Loïc’s disappearance of self.
That seems more human, and I think more Christian, than idealizing endless emotional martyrdom.
Where is Loïc after the miracle?
Once Charles returns, this became the question I could not stop asking:
Where is Loïc now?
Félicie sees Charles on the bus. She leaves. Charles follows. He learns about Élise. He has not forgotten Félicie. The woman who was travelling with him is not his wife. The extraordinary mistake of five years earlier is suddenly repaired. The family celebrates. But where is Loïc? Is he at home? Does he know yet? Will Félicie tell him immediately? Does he spend New Year’s Eve alone? Will somebody eventually love him in the way he wanted Félicie to love him?
The film does not answer. And I increasingly think that it is important not to repair that discomfort by inventing a sentimental continuation such as “surely God has someone better for him.” Maybe he eventually finds somebody. Maybe he does not for a long time. Nothing in the film guarantees distributive romantic justice.
Loïc did not lose Félicie because he failed a moral test. He simply was not the person she loved absolutely. That is moral luck in one of its most painful forms.
One person’s miracle does not settle everyone else’s story
This is where my theological reading moved beyond the question “Was Félicie’s faith rewarded?” I started asking:
What kind of theology of restoration is satisfied as soon as the protagonist receives what she wanted?
Shakespeare’s The Winter’s Tale itself should warn us against such simplicity. Hermione returns and Perdita is restored, but sixteen lost years have not ceased to exist. Leontes’s repentance does not rewind time. Mamillius does not return. Restoration changes the meaning of loss without deleting every consequence of loss (Shakespeare, n.d./2016).
That became the basis for one of the strongest conclusions I reached:
Resurrection cannot mean that one person’s restoration makes everyone else’s loss disappear.
The lost one returns. Good. Now ask who remains wounded. That question turns theology away from the winner’s perspective.
Élise’s tear complicates the happy ending
Élise makes the ending even stranger. Félicie cries in Charles’s arms and insists that these are tears of joy. Élise enters and looks at her parents. Then she leaves them, moves into the other room, sits with her toy and dreams. A tear runs down her cheek. Her grandmother asks whether she is happy to have recovered her father and tells her not to cry.
Élise answers:
« Je pleure de joie. »
Criterion reads the repetition as a Pascalian grace note, connecting the child’s phrase with the language of religious experience (Smith, 2024). That may be part of Rohmer’s design. But I would preserve the scene’s psychological ambiguity.
We do not actually know that Élise’s interior state is simple joy. It could be joy. It could include confusion. It could include displacement as the entire structure of her known family changes within minutes. It could be an emotion she does not yet know how to name, followed by repetition of the phrase she has just heard adults use. The screenplay gives us a tear and a sentence. It does not give us direct access to the child’s interiority.
This is strangely parallel to the larger epistemological problem of the film. Naming an experience does not completely settle what the experience is. Félicie calls something faith. Élise calls something joy. The words matter. They do not eliminate interpretation.
Rohmer controls the bus doors and reality does not
The final reunion is where my epistemological disagreement with the film becomes clearest. Félicie boards a bus. Charles happens to be there. They recognize one another. She gets off. He responds quickly enough to follow. The physical timing of doors, movement, recognition and decision all permit the reunion. Rohmer can make this happen because Rohmer has root access to the universe of the film. Reality generally does not provide us with administrator privileges.
This gave me the principle I kept returning to:
Narrative vindication is not epistemic justification.
A proposition can become true even if the evidence that originally supported it was weak. Suppose somebody predicts an extremely improbable event on insufficient evidence and the event nevertheless happens. The successful outcome does not automatically prove that the earlier confidence level was justified.
In probabilistic language:
Outcome = true does not imply Prior confidence = well calibrated
Rohmer’s story selects the Félicie whose Charles actually returns. Imagine instead one million people maintaining comparably improbable romantic hopes. One eventually meets the lost lover on a bus. A film made only about that person will make the strategy appear prophetic. The unresolved cases vanish from the dataset. That is a selection problem. It does not invalidate the beauty of the exceptional case. It prevents the exceptional case from becoming a general epistemology.
The film knows it is a fairy tale
To be fair to Rohmer, the film does not pretend its ending is an ordinary statistical expectation. Its Shakespearean framing tells us from the beginning that we are entering the logic of a tale, and La Cinémathèque explicitly describes the Shakespearean magic that eventually operates inside the grey winter world (Jamet, 2019). Another OpenEdition study emphasizes that A Tale of Winter uniquely turns winter into a symbolic condition oriented toward the “resurrection” of a lost summer (Herpe, 2013).
Most of the film remains aggressively ordinary: public transport, hair salons, suburban houses, family meals, awkward conversations, childcare, arguments, books, Nevers. Then one coincidence arrives carrying almost the full weight of fairy tale. That is why it works aesthetically. The improbable is surrounded by enough ordinary reality that when it finally enters, the world does not feel completely fantastical.
I can therefore say two things simultaneously:
The ending is beautiful.
and:
The ending does not prove that Félicie’s earlier certainty was epistemically justified.
I see no contradiction there.
Authorial providence is different from theological providence
This also led me to distinguish several meanings of “providence.” Inside the film there is authorial providence. Rohmer literally controls the sequence. He chooses the bus, Charles, the route, the timing, the emotional availability and the doors. Then there is theological providence, which concerns the interpretation of reality under divine care. And then there is ordinary contingency, in which events occur through secondary causes, choices, physical mechanisms and statistical coincidence. Those categories should not simply be collapsed.
A person may receive an improbable event with gratitude and interpret it as providential without claiming to have scientifically established that God directly altered a transport timetable. I think that is an intellectually healthy form of religious interpretation. Providence need not look phenomenologically different from chance. But that is a hermeneutic claim about meaning, not an experimentally demonstrated mechanism. The bus door is already doing enough work without asking it to solve the entire doctrine of divine action.
Félicie’s hope and Christian hope are not identical
Theological vocabulary becomes especially important here because the film repeatedly transposes religious categories into erotic life. The Catechism defines theological hope in relation to the Kingdom of heaven, eternal life, Christ’s promises and divine grace. It also says that theological hope takes up ordinary human hopes and purifies them, ordering them toward a greater good and away from selfishness (Catholic Church, 1992, §§1817–1819).
This makes an important distinction possible: “I hope Charles returns” is a genuine human hope. It is not, simply by itself, the theological virtue of hope. Charles is a finite human being. He cannot become the final object of salvation without being asked to bear a metaphysical load no lover can bear.
This does not diminish romantic love. In fact it may protect it. I can love a human being immensely without requiring that person’s continued presence to guarantee that reality itself remains meaningful. I can desire reunion without declaring reunion metaphysically necessary. I can mourn absence without making the absent beloved into a private eschaton.
Why Moltmann helped me become more critical of passive hope
Jürgen Moltmann’s Theology of Hope gave me a language for one of my strongest instincts while watching the film. Moltmann famously describes Christianity as eschatological from beginning to end: forward-looking, forward-moving, and therefore capable of transforming the present. His project explicitly concerns the responsible exercise of hope “in thought and action in the world” (Moltmann, 1967/1993).
This matters because hope can easily become confused with psychological endurance:
“I continue believing that what I want will happen.”
Moltmann’s account is more demanding. Hope changes one’s relation to the present. It produces unrest against what is treated as final. It opens action toward a promised future (Moltmann, 1967/1993).
That gave me a very practical question:
What does the hope make you do?
I originally used that question too aggressively against Félicie, because I had overlooked her earlier search efforts. Once those were restored, the picture became fairer. She acted when meaningful search actions existed. Yet the question remains useful later. When active search becomes almost impossible, her hope largely changes the way she manages commitments. It does not become nothing, but its practical form narrows. The ethical question then becomes whether preserving that hope is compatible with justice toward the people who inhabit the present with her.
Presumption may be the theological danger closer to Félicie’s certainty
Moltmann also helped me think about the distinction between hope and presumption. Hope reaches toward fulfilment it does not yet possess. Presumption behaves as though the fulfilment has already been secured. This is where Félicie’s language sometimes troubles me most. She knows, intellectually, that Charles may be married or no longer love her. Yet her deeper rhetoric often treats the bond as though its ontological permanence exceeds ordinary uncertainty.
Hope can say:
“I do not know, but I remain open.”
Presumption begins to sound more like:
“I know in a way that no contrary evidence can meaningfully challenge.”
The cathedral’s “I saw that I was not mistaken” approaches that boundary. Again, Rohmer ultimately makes her right. But being made right by an author is not the same as having possessed enough evidence to know.
Benedict XVI gives me the formulation I trust more
Spe salvi contains a line that became extremely important to my own understanding of hope. Benedict XVI describes serious and upright human conduct as “hope in action.” Human beings pursue lesser and greater hopes, work toward futures they value, and open possibilities even though the ultimate ground of Christian hope exceeds what human effort can guarantee (Benedict XVI, 2007, §35).
This gave me the formulation I now find much more convincing than passive optimism:
Hope begins where responsible control ends, not where responsible action is abandoned.
That sentence requires careful interpretation. It does not mean hope begins only after every conceivable action has been exhausted in obsessive fashion. It means that hope and agency are not enemies.
If something matters, I can investigate, act, seek help, reduce preventable risks, use the best information available and create fallback paths. Eventually I may reach a boundary where I cannot control the result. That boundary is not evidence that action was useless. It is the point at which action must coexist with uncertainty. This is the form of hope I find much easier to trust.
Hope can be intense without pretending to be knowledge
This also corrected another possible mistake: a rigorous theology of hope does not require emotional detachment. I do not have to say:
“Because I cannot know the future, I should avoid wanting anything strongly.”
That would turn epistemic humility into emotional anaesthesia.
I can want a good with enormous intensity. I can work for it. I can pray for it. I can fear losing it. And still say:
“I do not know whether it will happen.”
Formally:
intensity(desire) ≠ probability(outcome)
But if the hoped-for good actually occurs, humility does not prohibit joy. Quite the opposite. Because I did not regard the outcome as my possession in advance, I can receive it as gift rather than confirmation of my omniscience. This is where I finally found a version of the miracle in Rohmer that I can embrace without adopting Félicie’s epistemology.
The harder theological question is who suffers while I hope
Catholic hope also refuses to remain a private psychological asset. Hope is ordered toward charity. Prudence discerns the true good and appropriate means; justice concerns giving the other person what is due (Catholic Church, 1992, §§1806–1807). This changes the question.
Instead of asking only:
“Was Félicie’s hope eventually fulfilled?”
I also want to ask:
Who suffered while she hoped?
That question is not hostile to hope. It is what prevents hope from becoming egocentric. A hope that requires me to stop seeing my neighbour has become spiritually suspicious, even if the desired event later occurs. That is why Loïc matters so much to me.
Félicie is not evil but sincerity is not a complete moral defence
I want to resist caricaturing her. Félicie is not a calculating villain. She does not seem to enjoy making Loïc suffer. She does not seem delighted by Maxence’s pain. She loves Élise. She can recognize goodness. She admits that she is demanding. She feels shame when she realizes how her decisions affect Maxence. She is often very sincere.
But sincerity is only one virtue. A person can sincerely express an emotion while remaining unjust in the way that emotion is imposed on others. A person can honestly describe a desire while failing to perceive the system of care that makes pursuing the desire possible. This is why I eventually described Félicie less as hypocritical than as morally incomplete.
Her moral vision is highly developed along the axis of authenticity:
Am I living according to what I genuinely love?
It is less developed along another axis:
What is my way of preserving that authenticity asking other people to endure?
The second question is where Loïc enters.
The reliable person can become invisible infrastructure
There is a broader relational insight here that goes far beyond this film. The more reliably somebody carries weight, the easier it is to forget that carrying has a cost. A person who consistently answers becomes “the person who is always available.” A person who remains calm becomes “the person who can handle it.” A person who forgives becomes “the person who will understand.” Eventually a virtue can become infrastructure.
This is morally dangerous because infrastructure is precisely what we stop noticing when it works. Loïc’s capacity for generosity can therefore become the condition that allows Félicie’s instability to continue with lower immediate cost. Another person’s virtue can unintentionally subsidize my irresponsibility.
That insight is one reason I no longer think the question “Does Félicie have the right to choose freely?” is enough. Of course she does. The next question is what she does with the people who make that freedom survivable.
What would justice require if romantic love cannot be required?
Justice cannot require Félicie to manufacture desire for Loïc. So what might it require? I think the answer concerns recognition. She can recognize:
- that his availability is costly;
- that continuing to use romantic intimacy while denying romantic possibility can become cruel;
- that he may need distance;
- that his generosity does not make him infinitely durable;
- and that her happiness does not erase his grief.
In other words, reciprocity need not mean symmetrical emotion. It can mean mutual recognition of personhood and cost. This is much closer to what I want from an ethics of love.
Why I cannot simply condemn Félicie either
And yet, after all this criticism, I still cannot simply turn against her. She has one insight that I think is genuinely strong: choosing somebody permanently because he is good, safe and available is not automatically more ethical than refusing him. Loïc’s goodness cannot compensate for an erotic reality that does not exist. Maxence’s practical suitability cannot turn limited love into absolute love by administrative decision.
If Félicie married or permanently committed herself to one of them merely because probability said Charles was unlikely to return, she might create an even more serious injustice. This is why her refusal of false permanence deserves respect. The problem lies less in refusal itself than in the relational architecture surrounding the refusal.
Can I refuse possession without turning others into provisional resources? Can I remain faithful to an unresolved love while telling the truth about what I can and cannot offer people in the present? Can freedom become responsible without ceasing to be freedom? Those questions are much harder than declaring Félicie either selfish or saintly.
The million-Félicies problem
One thought experiment continued to bother me because it exposes exactly what narrative selection can hide. Imagine one million Félicies. Each has lost an extraordinary love through some accident. Each preserves hope for years. Each avoids permanent attachment partly because of the possibility of reunion. In one case, the missing beloved appears on a bus. In the other 999,999 cases, he does not.
Which story becomes a Rohmer film? Obviously the bus. And once the camera selects that life, the earlier hope acquires retrospective radiance. This is not a criticism of storytelling. Stories necessarily select. It is a warning against using narrative success as a probability model. The rare event may be real. It may even be beautiful beyond measure. It remains rare.
The miracle is morally incomplete if the camera stops too soon
By the end I realized that my strongest criticism of the film was no longer epistemological. It was moral and theological. Suppose I grant everything. Suppose Félicie’s hope is admirable. Suppose the bus encounter can be interpreted as grace. Suppose Shakespeare has correctly prepared us for restoration. Suppose the Pascalian wager has been narratively vindicated. There is still a question:
Does grace become complete simply because the person whose perspective we have followed is happy?
I do not think so.
If Christian hope culminates in reconciliation, then fulfilled hope should enlarge the field of concern, not shrink it. The camera should be able to look toward Loïc. And Maxence. And Élise. And the years that cannot be returned. Restoration is not deletion.
This is where the film finally became more beautiful to me
Strangely, I think I love A Tale of Winter more after refusing several of its easiest interpretations. I do not need Félicie to be a saint. I do not need her certainty to be perfectly calibrated. I do not need Charles’s return to prove that inward conviction functions as prophecy. I do not need Loïc’s suffering to be secretly compensated off-screen. I do not need Élise’s tear to have only one emotional meaning. I do not need providence to be detectable as a violation of ordinary causality.
The film is richer when all those tensions remain active. Rohmer gives Félicie the miracle. I can rejoice with her. But I can also ask what the miracle does not solve.
What Loïc finally taught me
I began the film thinking primarily about Félicie’s faith. I ended thinking about Loïc’s presence. He does something much less spectacular than seeing the future. He stays. He listens. He helps. He thinks seriously about what Félicie says even when he disagrees. He suffers jealousy without turning it into ownership. He prays for a happiness that may exclude him.
And yet the lesson I take from him is not that love means accepting endless pain. It is that love becomes morally serious when it survives contact with another person’s vulnerability without abolishing the giver’s own personhood. That is why I keep returning to:
Loïc’s generosity without Loïc’s disappearance of self.
It may be the most convincing form of love available anywhere in the film.
Hope, faith and certainty are finally three different things
After all this, I would distinguish the three words in my title very carefully. Certainty concerns how strongly I experience something as true. Certainty can be psychologically authentic and still mistaken. Human hope concerns a desired future whose fulfilment I do not possess. It can be rational or irrational, active or passive, generous or egocentric, depending on how it relates to evidence and action. Theological hope is something larger. In Christian terms, its final object is not Charles or any other finite desired outcome. It is grounded in God and therefore can take up finite hopes without making any single finite good absolute (Catholic Church, 1992).
Faith likewise cannot simply mean “I feel absolutely sure.” Christian faith contains trust, but trust is not the same thing as epistemic overconfidence. This is why Félicie fascinates me. She moves constantly across the borders among all three categories. Sometimes she hopes. Sometimes she claims certainty. Sometimes she interprets that certainty religiously. Sometimes Loïc supplies theological language she did not originally use. Sometimes the film itself confirms what reality had not yet confirmed.
The formulation I finally trust
The theology that emerged for me from the film can be stated quite simply, although it took a surprisingly long journey to reach it:
I hope for a future I cannot control. Because I hope, I act faithfully in the present. Because I do not control the future, I remain humble about what I know. And because Christian hope is inseparable from love, I cannot treat other people’s suffering as merely the price of my own fulfilment.
From that follow several principles I now find difficult to separate:
Freedom does not cancel relational responsibility.
Care does not create ownership, but care is not morally meaningless.
Subjective certainty is not external proof.
Hope begins where responsible control ends, not where responsible action is abandoned.
Grace cannot erase justice.
Resurrection cannot mean that one person’s restoration makes everyone else’s loss disappear.
None of these sentences requires me to reject Rohmer’s fairy tale. They simply prevent me from stopping my thinking when the bus door opens.
The question I still cannot leave behind
The traditional emotional question of A Tale of Winter is whether Charles will return. Rohmer answers yes. The more interesting question for me became what happens after that yes. What happens to the moral world that existed before the miracle? What becomes of the people who loved without being chosen? What does the child understand? Which relationships have been restored, and which have merely been displaced? Can fulfilled hope remain attentive to those who did not receive fulfilment?
This is where the title of the film finally acquired a different meaning for me. Shakespeare’s winter does not simply end because one lost person returns. Christian resurrection is not an editing operation that deletes the wounds preceding it. If restoration is truly restoration, it must eventually become large enough to remember what happened to everybody.
So when Charles appears and Félicie receives the extraordinary future she had preserved for five years, I can be happy for her. I can even call the moment miraculous in the narrative sense, and perhaps providential in a theological interpretation that remains humble about causality. But somewhere outside the celebration is Loïc. And that is precisely why I cannot stop at the miracle.
For me, the final theological question of A Tale of Winter is no longer simply:
“Can the lost one return?”
It is:
When the lost one returns, who is now lost?
Rohmer gives no clean answer.
References
- Rohmer, É. (Director). (1992). Conte d’hiver [A Tale of Winter]. Les Films du Losange / C.E.R.
- Criterion Collection. (n.d.). A Tale of Winter (1992). The Criterion Collection. https://www.criterion.com/films/29649-a-tale-of-winter
- Smith, I. S. (2024, February 13). Eric Rohmer’s Tales of the Four Seasons — Another Year. The Criterion Collection. https://www.criterion.com/current/posts/8387-eric-rohmer-s-tales-of-the-four-seasons-another-year
- Jamet, X. (2019, January 9). Éric Rohmer, mode d’emploi. La Cinémathèque française. https://www.cinematheque.fr/article/1357.html
- Andrew, G. (2015). Eric Rohmer original film posters. British Film Institute. The BFI describes A Tale of Winter as a study in “reason and responsibility, fidelity and faith.” https://www.bfi.org.uk/features/eric-rohmer-original-film-posters
- Shakespeare, W. (n.d./2016). The Winter’s Tale, Act 5, Scene 3. Folger Shakespeare Library. https://www.folger.edu/explore/shakespeares-works/the-winters-tale/read/5/3/
- Amiel, V. (2013). Le cinéma d’Éric Rohmer ou la transcendance comme un jeu. In Rohmer en perspectives. Presses universitaires de Paris Nanterre / OpenEdition Books. https://books.openedition.org/pupo/40162
- Herpe, N. (2013). Des Contes des quatre saisons à L’Anglaise et le Duc — La chute dans la parole. In Rohmer en perspectives. Presses universitaires de Paris Nanterre / OpenEdition Books. https://books.openedition.org/pupo/23630
- Moltmann, J. (1967/1993). Theology of Hope — On the Ground and the Implications of a Christian Eschatology. Fortress Press. https://books.google.com/books/about/Theology_of_Hope.html?id=sWPIcMupDeAC
- Catholic Church. (1992). Catechism of the Catholic Church — The judgment and formation of conscience, §§1777–1788. The Holy See. https://www.vatican.va/content/catechism/en/part_three/section_one/chapter_one/article_6/i_the_judgment_of_conscience.html
- Catholic Church. (1992). Catechism of the Catholic Church — The human virtues, especially §§1806–1807. The Holy See. https://www.vatican.va/content/catechism/en/part_three/section_one/chapter_one/article_7/i_the_human_virtues.html
- Catholic Church. (1992). Catechism of the Catholic Church — The theological virtues, especially §§1817–1819. The Holy See. https://www.vatican.va/content/catechism/en/part_three/section_one/chapter_one/article_7/ii_the_theological_virtues.html
- Benedict XVI. (2005, December 25). Deus caritas est, especially §7. The Holy See. https://www.vatican.va/content/benedict-xvi/en/encyclicals/documents/hf_ben-xvi_enc_20051225_deus-caritas-est.html
- Benedict XVI. (2007, November 30). Spe salvi, especially §§2, 35 and 41. The Holy See. https://www.vatican.va/content/benedict-xvi/en/encyclicals/documents/hf_ben-xvi_enc_20071130_spe-salvi.html