Stop being reasonable

People almost invariably arrive at their beliefs not on the basis of proof but on the basis of what they find attractive.

Monsieur Pascal has too much vacuum in his brain.

– René Descartes

HERE’S A PLAUSIBLE-SOUNDING RULE: CHANGE your mind when, and only when, there is more evidence for one side than the other. If the evidence does not resolve things, suspend judgment.

If that’s right, then it’s very strange that Nicole Kluemper changes her mind as often as she does. All the evidence is fixed and unchanging, sealed on videotapes and in evidence files in the California court district. But Nicole, professionals in psychology, and hundreds of people interested in her case are still divided over what really happened.

There are two parts to the story: what happened to Nicole when she was not yet seven, and what happened when she was seventeen. The first part begins when Nicole’s parents’ marriage soured into a bitter custody battle and Nicole’s father said Nicole had been abused by her mother. Nicole’s mother denied it, and the family court sent forensic psychiatrist David Corwin to break the impasse.

Dr David Corwin is a forensic and child psychiatrist specialising in child abuse. He chaired the group that founded the American Professional Society on the Abuse of Children in 1987, and by the time he was called to Nicole’s case he had already interviewed dozens of children in similar situations. Corwin was to conduct a series of interviews with Nicole and her family and then make a custody recommendation to the court. In the transcripts from those interviews, Nicole says her mother has hit her; her mother has hurt her in the bath-tub; her mother has burned her.

CORWIN: [Nicole], listen closely, did those things actually happen?

NICOLE: Yes.

CORWIN: Is that the truth?

NICOLE: Yes.

There were other troubling signs. Nicole’s father had a new girlfriend, who told Corwin she had seen burns, stamped into the soles of Nicole’s feet, that were shaped like the coils on a kitchen stove; she guessed that Nicole’s mother had held her daughter upright over a hot cooktop until her feet had blistered. Nicole seemed nervous when her mother sat outside Corwin’s office, and asked him anxiously whether his microphone could ‘talk’ to the waiting area where her mother sat. On separate occasions, Corwin asked each of Nicole’s parents to join their daughter in the interview room, and asked them each to instruct Nicole to tell the truth. Corwin paid close attention to how they reacted. ‘Just tell the truth, that’s what the doctor wants to hear,’ said her father, and his new girlfriend gave a slightly Instagrammish flourish: ‘The truth will set you free.’ But when it was Nicole’s mother’s turn, she said, ‘Do you remember what we talked about? What’d you tell Mommy?’

‘No,’ Corwin steers, ‘what I want you to tell her is to tell me the truth.’

Days later Nicole appeared in an emergency room with her mother and grandmother, this time saying that her father had abused her. But the next day, Nicole told a social worker she had made up the story because her mother had threatened to hit her if she didn’t.

Corwin scheduled one more last-minute interview with Nicole, the same day he was due to testify in court, to ask what had happened:

NICOLE: [Mom] threatened me that if I didn’t lie to the CPS that she would do something bad to me.

CORWIN: Are you a girl Scout or a Brownie or anything like that?

NICOLE: I’m a Brownie.

CORWIN: Okay. So what you just told me, do you promise on your oath as a Brownie, do you promise God that you’re telling the truth?

NICOLE: (Nods yes).

Nicole’s mother lost custody, and was not allowed visitation rights.

For ten years, that was that. But then there’s the second part of the story, the part that happened when Nicole was sixteen. I spoke to Dr Corwin from his office in Utah, where he still works in child forensic psychiatry and paediatric trauma, and asked him what he remembered of the events that came next. From pictures you might be tempted to compare him to Colonel Sanders if the colonel had become a doctor instead of going into fried chicken – he has a high forehead, a white moustache and a puffy halo of white-grey hair – but the comparison only sticks in pictures because when Corwin speaks he’s no jovial cartoon character. He is sober, precise and a little guarded with me, but it turns out the answer to the question about what he remembers is ‘a lot’.

‘I called [Nicole’s] father because I was being asked to speak to the American Academy of Pediatrics [on the topic] “Should we believe what children tell us about child abuse?”. I had thirteen minutes to address that topic – in front of 4000 paediatricians. And Nicole’s video was one of the most compelling descriptions of abuse I had seen in my professional experience. So I was seeking her father’s consent, and her assent, to show it in a big forum. I had sought their permission several years earlier to show it under controlled conditions to a sub-committee of the California Legislature. Her father and she had given me permission before, [but] it had been several years, and I felt that ethically I had a responsibility to make sure that they were still okay with it being used in forums of that type.’

Corwin tried to reach Nicole’s father via the old means, but discovered that since they’d last spoken he had had a stroke, and Nicole had been moved into foster care.

‘I ended up locating him in a convalescent home … He said yes, that he would confirm his consent, and I said, “Well, I’d like to speak to Nicole, too.” He gave me the number of the foster home. When I rang, I explained to her why I was calling, and said, “Is it still okay with you if I show that for professional training?” And she said, “Sure, Doctor Corwin, anything to help” – that was roughly what she said. So I’m feeling like I’ve done my ethical duty and was going to wish her well, and there’s this little pause and she says, “But you know something? I can’t remember what happened when I was little. Would you please send me the videos?” ’

Many reports of this moment have suggested that Nicole had wholesale forgotten alleging the abuse at all. That isn’t right, at least not going by the transcripts and her own recollections. She had forgotten the precise claims she’d made and, more importantly, she’d forgotten whether they were true. Sometimes she had vivid flashbacks to being held over a stove, but sometimes she had that foggy feeling we get when we can’t tell which memories are real and which we’ve stitched together out of fragmented anecdotes.

She had been worrying about it before Corwin phoned. After her father’s stroke, Nicole had reached out to her mother and they had just started speaking again for the first time since the divorce. She had hoped that some familial contact would help with her sudden sense of being adrift and alone. It did. With her foster mother’s encouragement, she started speaking to her mother more regularly. But it wasn’t long before her mother started saying to Nicole what she’d said to Corwin all those years ago: that the abuse story was a lie, that she had never hurt her daughter.

‘What if I just said it?’ Nicole started wondering aloud to her foster mother. ‘What if Dad put me up to it?’ When Corwin phoned, Nicole saw a chance to get the evidence that would help her make up her mind. He could just send her the tapes.

Corwin wasn’t sure how to respond. ‘To my knowledge that had never been done, never been studied,’ he told me. ‘We had no idea what the possible pluses and minuses are of a young woman looking at such material. I had two thoughts. One was [that] she has a right to know what happened in her life, and on the other hand I thought, “I certainly don’t want that to hurt her, or be upsetting to her.” So I said to her, “I think it’s your right [to see them], but can you wait until I can be there and I can do everything I can to make it a positive experience?” ’

About a year later Corwin happened to be near Nicole’s home town. Nicole’s father had died in the intervening time, so Corwin had to locate Nicole through her school, and when he did she told him yes, she still wanted to see the tapes. Corwin arranged to show them to her in the company of her foster mother and a local therapist, and with a video camera to record her informed consent.

CORWIN: Okay. So why don’t we start with – if you tell me what you remember saying.

NICOLE: Do you want little things, or –

CORWIN: Everything. Everything you can remember.

NICOLE: Okay. I remember visually, I remember wooden, like the walls, it was like a wooden paneling, and I remember in one of the interviews I wore a shirt that was striped this way … I remember I was answering questions, and I told you, I guess, I told the court, that my mom abused me, that she burned my feet on a stove, that’s really the most serious accusation against her that I remember, I don’t know what else I said.

CORWIN: Okay. Do you remember anything about the concerns about possible sexual abuse?

NICOLE: No, I mean, I remember that was part of the accusation but I don’t remember anything else – wait a minute, yeah I do.

CORWIN: What do you remember?

NICOLE: (Pauses) Oh my gosh, that’s really (closes eyes and holds eyes) really weird. I accused her of taking pictures (starts to cry and foster mother puts hand on shoulder) of me and my brother and selling them and I accused her of – when she was bathing me or whatever, hurting me, and that’s –

CORWIN: As you’re saying that to me, you remember having said those things or you remember having experienced those things?

NICOLE: I remember saying about the pictures, I remember it happening, that she hurt me.

CORWIN: Can you be more specific because I –

NICOLE: – I know what was said on the tape. On the tape it was said that she put her fingers in my vagina.

CORWIN: Okay. Is that what you recall, or –

NICOLE: That’s what I recall. I recall saying it, and I recall it happening.

It seemed to many in the room that Nicole’s memory had just given her the certainty she’d been looking for. It’s not clear where the story about the photographs comes from; that wasn’t on the original tapes. But her description of the bathtub incident matches what she said as a child, and broadly speaking so does the story about the burns. Nicole had this sudden recollection before seeing the tapes, but Corwin played them for her as planned anyway, and when they were finished Nicole’s foster mother said, ‘I think this has been a beautiful closure.’ Seventeen-year-old Nicole agreed:

NICOLE: I’m glad now that I don’t have to keep trying to convince myself that my dad ever lied to me, because for a while after he died I was convinced that I was going to have to convince myself that he lied, and told me those things about my mom. My mom, I – I do plan on continuing my relationship with my mom, because she is my mom, and I don’t want to not allow her to be a part of my life and then feel bad when she’s gone. If I put myself in her place I have to say that if I was ever going to do that to my child, my life would depend on the fact that they would give me another chance. So I have to give her another chance.

CORWIN: So despite its painfulness, do you think that this has been a constructive experience at this point in time?

NICOLE: Yes. There’s some questions that might never be answered, you know, I can live with that. My biggest question has been answered. I think it’s a very healthy thing not to run from something, just the fact that I turned around and I faced it, that’s strength enough to go on, to put it to bed. And next you’re going to ask if you can use this for educational purposes, right? (Laughs) Yes, you can.

With Nicole’s permission, Corwin began work on a journal article about the incident, calling Nicole ‘Jane Doe’ throughout. Nicole went back to her final years at school, believing that at last she had the evidence to avoid being uncertain.

Withholding belief has something of a brand reputation for being rationally responsible. Agnosticism in the face of indecisive evidence seems like the mature response, one that does justice to the complexity of the situation. As Clarence Darrow famously wrote, a few years after that defence of a school teacher accused of teaching evolution in a Tennessee school, ‘Any one who thinks is an agnostic about something, otherwise he must believe that he is possessed of all knowledge. And the proper place for such a person is in the mad-house or the home for the feeble-minded.’

The inverse is true, too; often when we see people leaping to form beliefs when the evidence is indecisive – like when people construe scant evidence from a crush as proof positive that their feelings are reciprocated – we can feel a kind of pity, or a vicarious shame. You should resist such shoddy conclusions when the evidence is indeterminate; you should keep your mind in a kind of suspended animation, floating above all the available hypotheses.

One path to this conclusion, based on a plausible view, goes like this: evidence, and only evidence, determines what we should believe. Change your mind when there is enough evidence, but only when there is enough. In cases where the evidence on both sides of the belief scale compels it to stay level, suspend judgment, since the only thing that could resolve such an impasse would be more evidence. Considerations like what would be good for us to believe ought not to be able to leap onto the scale and tilt it in either direction. Such considerations are not useful, and they just get muddy paw prints all over the instruments. David Hume’s formulation is one of the cleanest – ‘A wise man … proportions his belief to the evidence’ – and our friend William Clifford’s is one of most fiery: ‘It is wrong always, everywhere, and for anyone to believe anything on insufficient evidence.’

The question about whether other considerations should be allowed to pull us out of agnosticism is then either answered with a resounding ‘No’, or just dissolves into ambiguity over what we meant by ‘should’. Thomas Kelly, a Princeton philosopher who specialises in the ethics of belief, imagines out loud how a philosopher of a certain stripe might react to the question about what we ‘should’ believe:

In cases in which what it is epistemically rational to believe clearly diverges from what it is practically advantageous to believe, there is simply no genuine question about what one should believe: Although we can ask what one should believe from the epistemic perspective, and we can ask what one should believe from the practical perspective, there is no third question [about] what one should believe, all things considered.

Seventeenth-century mathematician and philosopher Blaise Pascal gave us a famously unpersuasive example of a rival thought. ‘Au contraire,’ said Pascal (I’m summarising for efficiency but keeping the authentic French feel). Pragmatic considerations can break us out of agnosticism, he suggested, and in at least one case they should. That case has been dubbed ‘Pascal’s Wager’ and it was the Catholic Pascal’s effort to turn agnostics into believers. Suppose God doesn’t exist: what harm does it do you to believe he does? Not a great deal – an error, but a harmless one, and you’re worm food by the time you discover it. But suppose God does exist: what harm does it do you to believe he doesn’t? Hell, wrath, posthumous pitchforks; ‘If you gain, you gain all; if you lose, you lose nothing,’ wrote Pascal. ‘Wager, then, without hesitation that He is.’

I know of no religious person persuaded by Pascal’s wager, not even Pascal himself. Part of his own conversion involved a midnight vision in 1654, long before he wrote the wager, which he saw so vividly immediately wrote every detail on a small slip of paper, and for the rest of his life he carefully stitched that piece of paper into the lining of his coat, unpicking it and transferring it every time he changed coats. That doesn’t look like someone weighing the pragmatic considerations to me; it looks like someone who thinks they have seen an anvil crash down on one side of the evidence scale.

Contemporary philosophy takes it as background that your theory of justified belief has jumped the tracks if it winds up endorsing anything like Pascal’s view. ‘We know that Pascal’s wager – even if it is a good one – does not make it easier to know that God exists,’ wrote Mark Schroeder, a University of Southern California professor more sympathetic to pragmatic influences than many. ‘Cases like Pascal’s have made it seem obvious to many philosophers over a long period of time that practical considerations are not the right kind of thing to be relevant.’

The widespread attitude in favour of suspending belief is nowhere so prevalent as in criminal law, where the point is to suspend belief until we can believe responsibly. After all, we can do real damage to real people if we don’t.

Just ask Elizabeth Loftus.

Elizabeth Loftus is a psychologist and professor at the University of California, Irvine, who has spent most of her career working on the fallibility of memory and the costs of getting it wrong. When she was a graduate student she found that people could be induced to ‘remember’ false features of a filmed car accident if they were asked ‘How fast was the car going when the accident smashed its headlight?’ instead of ‘Did you see any broken glass?’ As a professor, she led the famous ‘Lost in the Mall’ experiments, in which experimenters told participants a series of stories that they said were true, like ‘Your parents told us about the time you got lost in the mall’. Participants would ‘remember’ these events, and even embellish them with ‘recollected’ details – even if the story was entirely false.

For most of her career, Loftus has worried what damage might be wreaked on the basis of false memories that felt true to the people who held them. In the 1980s came a particularly vivid answer: it emerged that a number of therapists in the UK, Australia, the United States and Canada had been using ‘hypnosis’ and strong barbiturates to help their clients pull ‘memories’ of trauma from their subconscious hiding places, trading on the thought that many adult neuroses were expressions of otherwise completely forgotten trauma. Stories of cartoonish evil, like ‘Mum and Dad ate the pizza boy’, started being ‘recovered’ as genuine memories, accepted as fact, and used to send people to gaol. Loftus’s research helped vindicate the flabbergasted accused by proving that the evidence for these memories was as hard to find as the memories themselves had been. Around the world sentences were overturned and multi-million-dollar settlements reached as clients and therapists had to reckon with the idea that the trauma they had imagined was the root of their problems was just that: imagined. One woman who recovered the ‘memory’ that her family were cannibals carefully collected a sample from a family lunch, only to have the laboratory send back its nonplussed reply: just beef. By the time David Corwin and his co-author Erna Olafson published an article titled ‘Videotaped Discovery of a Reportedly Unrecallable Memory of Child Sexual Abuse’ in 1997, describing Nicole’s experience using the pseudonym ‘Jane Doe’, Loftus had already written the book on the myth of repressed memory. It’s called The Myth of Repressed Memory.

Loftus read the article about Nicole’s sudden recollection and wanted to investigate Nicole’s case for herself; she thought it sounded fishy. But the name ‘Jane Doe’ wasn’t much to go on, and all the other details that could have identified Nicole were obscured: Nicole’s parents lived in ‘Mumtown’ and ‘Dadtown’, the date of her father’s death had been changed, and her foster mother was not referred to by her name. Nicole had been a minor at the time her sudden recollection, and it had been important to Corwin and Olafson to protect her identity.

Loftus broke through that pseudonym.

She spoke to me about the case one evening as she sat in her living room waiting for dinner guests. You have to talk fast with Loftus: she is busy and you don’t scare her. Once, she got a series of death threats for her work on recovered memories; not everyone was delighted to have been proved wrong. So she learned to shoot a gun, and the target practice sheets are still hanging in her office.

‘How did you find Nicole’s real name?’ I ask.

‘The very first clue that I got was that one of the people that Corwin sent the tapes to – because they wrote a commentary in the original article – was a PhD student of mine, who showed me the tapes in my living room, and I heard Corwin slip, and he called her ‘Nicole’. Then I heard the name of her real home town. Those were my first clues. Then it was a process. She said her father had died eleven months ago, so I narrowed in on the death. Then I found a court case with some similar facts. Then I taught myself how to search social security records, to plug into the death records. Then I thought, “I know, I’ll write to the newspapers and see if I can get the obituaries, see if anyone left a daughter named Nicole.” It was a long process.’

When Loftus had the name of the Jane Doe in the case, she sent a private investigator to Nicole’s home town. Nicole found out that someone had discovered her identity when a friend paged her to ask her why private investigators were knocking on doors.

Here is what Loftus and her team found.

A friend of the family told Loftus’s team, ‘No way did any of the allegations occur.’ Nicole’s brother said, ‘No way did my mother ever abuse Nicole.’ Nicole’s mother, whose relationship with Nicole had been strained and occasional ever since she had seen the tapes as a seventeen-year-old, wept and threw herself around Loftus’s shoulders, saying, ‘I never thought this day would come.’ Loftus found concerns about how credible Nicole’s father’s testimony could have been: people told her he had drinking problems, and Nicole’s grandmother had written to the court to say he once beat his own son so that ‘his entire face was swollen to a pulp and he was unable to move’.

Loftus spoke to the woman who had been dating Nicole’s father at the time of the allegations, the one who said ‘the truth will set you free’, and had told Corwin she had seen the burn marks on the soles of Nicole’s feet. She told Loftus that she and Nicole’s father had tried to get Nicole away from her mother with ‘the sexual angle’. ‘We were going for broke,’ she said, and recounted taking Nicole to different hospitals ‘to get documentation’ of the injuries on her feet. Even the burns themselves were open to doubt: Loftus found a letter from another psychologist, saying ‘it was never determined if her feet and hand were indeed burned, since [Nicole] has a fungus condition that causes her skin to blister and peel’. Nicole’s mother has the same condition – she showed Loftus patches on her fingers where the skin was shedding and sloughing off. Loftus also found Nicole’s foster mother, who said that she had taken Nicole to visit the kitchen where the burns might have happened, and found that the stove was gas. No coils though of course, the stove could have changed. Then, the piece of evidence Loftus calls ‘most impressive’: she found reports by another therapist who had investigated the case before Corwin, who had said that Nicole sounded ‘not spontaneous’, and ‘mechanical and rehearsed’.

While Loftus was conducting these interviews, Nicole phoned around trying to find out who had sent the investigator. She was just out of high school, and she knew the only thing she’d been involved in that might interest an investigator was the incident with the tapes, so she phoned Corwin, distressed, and when they discovered Loftus’s name Nicole complained to Loftus’s university.

‘[University staff] basically came to my office and took my files,’ says Loftus. ‘I mean, I really couldn’t believe this, because I was pretty convinced that I had a case of somebody who was probably falsely accused – I was interested in another human being, and I was going to get in trouble for it? It was shocking. I didn’t publish the exposé until [two years later], when I got my neck out of the guillotine.’

Loftus and co-author Melvin Guyer wrote these findings in a two-part article for the Skeptical Inquirer titled ‘Who Abused Jane Doe?’ Part 1 concluded: ‘we believe that there are ample reasons to doubt whether Jane Doe was physically or sexually abused by her mother, and to doubt much of the “supporting evidence” used to support the abuse hypothesis’.

Surrounded as we are by the sorts of stories that Loftus’s career has focused on, it’s not hard to see why people wind up thinking that rationality favours withholding belief in cases of evidentiary dead heats. Errors have costs. Better to suspend belief and have the guarantee that we have not made a costly mistake. Better, as Loftus puts it, to ‘doubt the hypothesis’, than to risk prematurely believing something that turns out to be wrong.

But notice that if we take this line of thought, we’re no longer considering evidence and only evidence. The question has stopped being about whether the evidence has anointed a particular hypothesis the right one, and started being about whether there are other good reasons for refusing to hold either hypothesis – good reasons like the cost we would have to answer for if we believed it and turned out to be wrong. If certain errors are particularly costly, or if they outweigh the costs of withholding belief, then it’s a short step to the conclusion that rationality recommends withholding belief in those cases. Mark Schroeder has argued along these lines, to the effect that certain arrangements of error-costs can mean that ‘having better evidence’ for something ‘can still fail to make it rational to believe’: that pragmatic considerations might yet turn out to have their place in determining what it’s rational to believe, despite the bad name they got from Pascal.

And if costs of error matter to us, then we might join William James in thinking that ‘believing truth’ and ‘shunning error’ turn out to be two materially different missions, with opposing recommendations about what to do when the evidence is indecisive. ‘We may regard the chase for truth as paramount, and the avoidance of error as secondary,’ James explains, ‘or we may, on the other hand, treat the avoidance of error as more imperative, and let truth take its chance.’ As Thomas Kelly formulates it:

The more weight one gives to the goal of not believing false propositions, the more it behooves one to be very conservative about what one believes, and to believe only those propositions for which one has a great deal of evidence. On the other hand, the more weight one gives to the believing of true propositions, the more it behooves one to believe large numbers of propositions, including propositions for which one does not have a great deal of evidence.

So which should we choose?

‘What would you believe, if you were Nicole?’ I ask Loftus towards the end of our conversation.

‘Well … the father has passed away, and the step-mother doesn’t seem to be a part of her life, and one hypothesis about what happened was that they were a force in getting her to believe this … I don’t know what her relationship with her biological mother is now – if she has a relationship with the biological mother that would be a pull maybe to not believe it, but if she has a relationship with some of the repression aficionados who might want to keep her believing, they may be an influence. I just don’t know. It would be interesting to know. People would wonder, “Where is [Nicole] today? What is she thinking today?”’

‘Some days I believe it happened, and other days I don’t. I still go back and forth. The majority of days I just accept that it happened, but … there are still days when I don’t believe it.’

Nicole is a grown woman now, with children of her own. I realised just before I spoke to her that I had no idea what sort of person she might be. So much of the material you read about her lets her actual character recede into folds and folds of plot points. First she is literally nameless, a ‘Jane Doe’ you aren’t meant to be able to picture. Then she is pushed and pulled through child protection systems where lawyers and psychiatrists and foster carers speak on her behalf because legally she can’t do it alone. When she finally acts under her own name and volition by complaining to Loftus’ university, it is because hundreds of people have become interested in a story where she features mainly as a mute figure. She is the person we do not hear from, who facilitates the plot but who we don’t know much about, like a parent killed off in the first chapters of a Roald Dahl book.

Much of her life is exactly the opposite of this picture – it is firm, self-directed, knowable. She graduated in the top five per cent of her high school, enlisted with the US Navy when she was eighteen and became a helicopter pilot, flying pointy-nosed Apocalypse Now– looking behemoths with blacked-out windows and scary names like ‘Seahawk’. She was stationed in Hawaii, did routine military-grade fitness tests and won a Navy scholarship to go to university, where she received a BA in psychology. After she quit she earned three more degrees in psychology – two Masters and a PhD. She now practises in a clinic not far from where she grew up, and she is careful to tell me that what she says in our interview is not professional advice. She has the same mop of tight brunette curls she had as a child and the same wide, almost leonine face. Her cheeks stand out from the rest of her face in high relief and she has one deep dimple when she smiles, but ours is not a smiling conversation.

‘It was surreal to sit there and watch yourself knowing that it’s you. It’s still surreal, even when I watch those videos now. I had hoped that my whole memory would come back, that I could see the little girl in those tapes and it would be resolved once and for all. That didn’t happen. Actually for a while it did resolve it, for a brief period of time, and then all of the Loftus stuff threw it back into doubt.

‘I became a very angry person when Loftus came into my life,’ she continues. ‘There was a time, a long time actually, where I built my identity around being a victim of sexual abuse, and a victim of Elizabeth Loftus. I just [didn’t] understand how she could justify that.’

Nicole found out about Loftus’ article when it was already on newsstands. She and a friend from her Navy station went to a bookshop to buy a copy and read it together, shocked. Even though Loftus had reverted to the original ‘Jane Doe’ pseudonym, Nicole sued for defamation and invasion of privacy.

Part of the problem was that when Nicole read the article, she saw errors.

Loftus’s article wrongly claimed that Corwin had sought Nicole out with the express purpose of discovering ‘what, if anything’ Nicole had remembered. It wrongly stated that Corwin telephoned Nicole’s foster mother, ‘saying he was doing research and wanted to interview [Nicole] again’. It wrongly stated that Corwin believed the incident was proof of repressed memory. Corwin was unambiguous when he spoke to me: ‘We never claimed that we’d proven this or that,’ he said. ‘We just said, “This is interesting, and it’s presented here for discussion.” We didn’t take a strong position with regard to what it meant. We did raise the question that there are hundreds of thousands, maybe millions of these [forensic] videos in existence. What is the duty of the systems that maintain them? What are the rights of the children as they get older to know the basis of what happened during their childhoods? We asked that question. But no one’s touched that.’

Loftus’s article tendered Nicole’s brother’s remark that ‘no way did the abuse occur’ without mentioning that he had suffered a serious brain injury and been left with significant memory loss. It wrongly claimed, as Loftus did in a TED Talk in 2003, that Corwin had been showing the tapes ‘in public’. Corwin does not show them in public, and he would not send me a copy; ‘everyone’s wanted them’, he said, and ‘no one’s got them’. When he presented them at the California Legislature, the TV screens were turned so that the video was not visible from the gallery. And it seemed to matter that Loftus had not distinguished Nicole’s sudden recollection from those ghoulish ‘recovered memory’ stories. Hers was not a gnarled grotesque that had appeared for the first time on a therapist’s couch. She had said it as a child, and Corwin hadn’t specified which parent he was asking about in their meeting when she was seventeen. He just asked if she had remembered any sexual abuse and the scene had unfolded before her. These arguments shuttled back and forth for years during the defamation and invasion of privacy lawsuit, each one trailing clouds of footnotes. Years later, mutually exhausted, Loftus and Nicole reached a settlement.

For Loftus, none of those errors touched the substance of the article.

‘When the lawyers were trying to negotiate some sort of settlement, they said to me, “Is there anything that you can apologise for?” The only thing I could think of was that I was sorry that she felt so hurt by this effort, but I wasn’t sorry that I stepped into support what I thought – and now probably believe even more strongly – was a falsely accused woman … People want to say, “This is [Nicole’s] story, her rights, only her rights.” She’s not the only person in this story! How about the option that she realises that maybe, possibly, [the abuse] didn’t happen, that she reunites with her mother, and the mother has a daughter? That might have been a happy ending.’

‘Was that what you hoped would happen?’ I ask.

‘I might have even had a little fantasy that that might happen.’

But for Nicole, all that back-and-forth of evidence made the question of what to believe even more vexed.

‘I still haven’t found one piece of evidence that sways me one way or the other,’ she says. Instead, Nicole’s mind has changed for years like a weathervane in a hurricane.

‘When I watch [the videos] it’s very difficult to believe that that little girl is lying. So when I think about that, I’m almost unanimously persuaded that it did happen.’

‘How do you feel about the girl in those tapes on the days when you believe it didn’t happen?’

‘I don’t know if I even have an answer to that question. I mean, she believed what she was saying, but I was either coerced or tricked into believing that that was true.’

‘It’s interesting that you talk about those videos in the third person, like “that girl” and “that person”. Does it not feel like it’s you?’

‘At times it doesn’t. I think I talk about it more in the third person when I’m talking about it not having happened because it’s hard for me to own that part of myself, if I’m believing that I lied or was somehow not truthful in what I was saying.’

‘What does it do to a person, to have to withhold judgment about something this big?’

Nicole is quiet for so long I think the line might have gone dead.

‘That uncertainty … it was the ether that I lived in. It was all around me, all the time. It affected every single relationship, in every possible way. It requires me to have a sense of self that is not dependent on whether I was sexually assaulted by my mother. It’s a really big ask.’

In some ways Nicole’s story looks like a story about the costs and pragmatic consequences of various beliefs. She told me that those things weigh on her as much as the evidence itself.

‘On the days when I believe that it happened, I know that I am willing myself to believe that it’s true because there are people out there who believe strongly that it happened, and because I don’t want Elizabeth Loftus to be right. Going the other way, it takes a little less work, but I know I’m willing myself.’

In other ways, though, her story is about more than costs of belief: it is about the costs of withholding, too. If costs of error matter to us, why not the costs of withholding?

Believing something false is not the only way to cause monstrous shockwaves and incalculable damage. Failing to believe something that turns out to be true can be costly in similar ways, or at least to similar degrees.

Nicole’s mother is alive, for instance. Should Nicole speak to her, or not? Suppose her mother is innocent and, instead of believing that, Nicole spends her life suspending judgment. Are the costs Nicole – and her mother – would have to bear any better than the costs they would bear if she falsely believed ‘She did it’? I’m not sure they are: the actions of a person uncertain about whether their mother abused them can look very much like the actions of a person who positively believes she did. In neither case will their mother be entirely trusted with the grandkids; in neither case will she be vindicated; in neither case will their relationship be free of the strain of suspicion.

The problem here is that it’s difficult to find an action as lukewarm as the mental state of a person who isn’t sure what to believe. In the domain of belief, typically we have three options: believe the thing, reject it, or suspend judgment. But as Richard Feldman, a philosopher whose work on rationality is so prolific they gave him his own conference called ‘Feldmania’, notes: when it comes to actions, sometimes we find ourselves down to two options: do the thing, or don’t. ‘Hesitating forever is the same as not doing it,’ Feldman noted. Speak to your mother, or don’t.

Nicole does speak to her mother, but it is not an ordinary relationship.

‘There was a conversation not too terribly long ago when she said she was going to write out her justification for how I should know that this didn’t happen, but I’ve yet to see that and I don’t even know if I want to. I just don’t believe that she could convince me one way or the other.’

‘Was there ever a time where you just … got on easily?’

‘There have been times like that, but rather quickly it seems to come back to … she wants me to be a seven-year-old. She doesn’t want me to be a forty-year-old. I’ll give you an example: on my thirty-seventh birthday she threw me a seven-year-old’s party,’ Nicole recalls. ‘The cake had to have my favourite flowers on it, she sent one of my friends to the store to get ice cream, it was just … I realised in that moment that I could just never give her what she wants from me, to go back in time and be allowed to mother me again. It’s … it’s a very strained relationship.’

If the costs of error matter, why not these costs, too: the costs of withholding belief? Some philosophers have answered simply: ‘they do’. Mark Schroeder, for instance, has argued for a distinction between the error of believing falsely and the error of missing out on truth by believing nothing at all, and argues that both wield some sort of influence over what we ought to believe: ‘plausibly, one important source of reasons to withhold [belief] will come from the preponderance of the cost of having a false belief over the cost of missing out on having a true belief’.

One of the weirder consequences of thinking that costs matter, though, is that we will wind up with the verdict that different people facing the same evidence ought to believe different things depending on the stakes and costs around them. Maybe this seems a little skew-whiff to you; maybe you don’t like the thought that what it’s rational to do could depend on features about a given thinker. Perhaps you’re sympathetic to Tyler Burge’s thought in his work on justifications for and entitlement to beliefs: ‘Reason has a function in providing guidance to truth, in presenting and promoting truth without regard to individual interest. That is why [reasons for belief] are not relativized to a person or to a desire.’

Whatever the aims of rationality and the status of costs of error, Nicole’s story gives us cause to rethink our confidence that reason favours withholding belief in all cases where the evidence is indecisive. It may yet turn out that that view is right, but it will be for far more complex reasons than we have in mind when we act the way I said we do at the start of this story, as we look with a mixture of pity and vicarious shame on people who leap to believe their crush reciprocates their feelings. Even if our reaction turns out to be justified, we will still have important unanswered questions about what rationality aims at, and about whether costs matter, and why.

Unable to resolve what to think, Nicole has resolved what to think about. After a lot of therapy she has reached a rickety version of equilibrium: instead of trying to find her way out of uncertainty, she finds ways out of its costs.

‘I have to sort of build my life, my sense of self around being a wife, and a Christian, and a psychologist, and a friend. Those are the parts of my identity I build on. It’s a matter of making the conscious choice over and over again, until it’s a habit, to consider other things and other parts of myself. In a way it’s healthy, I think. Some people are able to build on their identities as survivors in a very healthy way, and then some people absolutely are not. When I finally made the choice to start living otherwise, it’s just like dieting: you start, and then you lose your motivation and you go back to the way you used to eat. I would start, and then I would revert back to my old way of thinking. All told it probably took several years for me to adjust completely.’

‘Do you have anything you’d say to people who are having to live with really heavy uncertainty?’ I ask.

‘That’s a really tough one,’ Nicole answers, after thinking for a while. ‘It’s just like forgiveness, in that everyone has to do it in their own time. If you try to rush someone they’re just going to dig their heels in and say, “Leave me alone.” Everyone has to make a decision for themselves as to when it’s time to move on. I didn’t want to get to the end of my life and know that I lived my entire life in fear. I’m not saying it will happen overnight, but … it can get easier.’

As I researched Nicole’s story, over and over, people would ask me, ‘So did it happen?’ Uncertainty, whatever its costs, is deeply unsatisfying. But there isn’t an answer – or, rather, there is, and we don’t know it. All we have is the evidence that both sides say matters most – and one other piece of information I found in reporter Michael Hall’s several-part investigation of a separate case for The Texas Monthly. Years after Nicole had her second meeting with Corwin, her foster mother – the one who had said ‘this has been a beautiful closure’ – turned up at a Texas police station claiming that the children she was fostering had been drugged and forced to dance at a ‘swingers club’ where adults watched and filmed them. One of the children who was supposed to have been there later recanted. ‘It never happened,’ she said.

Does that matter? This foster mother wasn’t in Nicole’s life when she first made the allegations, as a child. You decide what to believe.