Say you're most of an hour into the recording. Your grandmother stops, looks at the recorder on the table, and says that this next part stays between the two of you. And you say, of course.
It's the kind thing to say. Whether it's an honest thing to say is what the next twenty minutes are about, so hold off saying it.
The last lecture met this moment and gave you two ways with it. One was to switch the recorder off. The other came from Willa Baum, the early oral historian: keep it running, and promise to cut that part afterwards if she asks. It left the weighing of promises like those to this lecture.
So what does she have a right to, and what can you honestly promise her?
First, what you owe her before the recorder goes on. Then a university that made your promise in writing, to people with far more to lose than your grandmother, and what the courts did with it. Then what, if anything, belongs on paper, which is where the law comes in. And last a grandmother who was glad to be recorded, until she read what her granddaughter made of it.
Start an hour before that moment in the kitchen, with the recorder still off. What would you have had to tell her by then?
The third lecture had the case of telling a narrator nothing: a woman in Charleston who took her caller for someone from the Welfare office, and a caller, Jessie Butler, who let her.
The Oral History Association has a name for what that woman was owed. Informed consent is her yes, given when she knows what she's saying yes to, and put on record. What follows is mostly its account of it, as it would have gone for your grandmother.
She knows what the recording is for, and who is going to hear it, down to whether any of it goes online. She knows that what you intend for it is a plan, and no guarantee. She knows she can refuse a question, or the recorder, or all of it. And she knows what happens afterwards. The Association strongly recommends that a narrator be given the chance to review an interview and approve it before it's used, and among the choices it gives her are taking parts out, and keeping the interview closed until a set date. Its words for that review are whenever possible. For a family I'd make it a promise: she hears it before anyone else does.
A word for each of those choices. A part that's cut is gone for good. A part that's closed still exists, and the agreement is that nobody hears it until a date she names.
Had she known all that, she'd have been choosing in that kitchen. As it was, she was hoping.
There's an obvious thing to say back. She's your grandmother. She'd agree to anything you asked, and consent sounds like paperwork for strangers.
Janet Finch was a researcher in England, and she was a clergyman's wife. The people she interviewed were clergymen's wives. They often told her very private things, and what they had in return was her word that she'd disguise them. She wrote that she came away from interviews feeling that "my interviewees need to know how to protect themselves from people like me."
In your kitchen, you are people like her. What I take from Finch is that the narrator who trusts you most is the one least likely to ask what becomes of the recording, so the terms are yours to bring up.
Then get her yes onto the recording, in her own words. When the recorder goes on, and you've said who is asking, don't hand her a question that only wants a yes. Ask her to say what the two of you are doing, and what she wants done with it. You might start her off like this. Before we begin, tell me in your own words what we're doing today, and who you'd like to hear it. Then let her say it. Keeping the recording is one thing, passing it round the family is another, and putting it online is a third, and her yes should say which. Whoever hears it later then knows what she agreed to. Anyone else whose voice will be on it is asked as well.
That yes is harder when her memory is failing. The Alzheimer's Society, a different body from the association in the last lecture, has guidance on it, written for people doing research with people who have dementia. It starts from the assumption that she can decide for herself, and has the researcher consider whether she can. Her ability to agree may come and go, it says, so it treats consent as a process and not as one moment. Its last step is to stop, before or during, if she no longer wants to take part or is no longer able to, going by what you see as well as what she says. I'd also check with her again at the start of each session.
The Society's first step is to go through the people who look after her. I'd ask them how she is, and what she'd have wanted. While she can decide for herself they don't stand in for her, and I wouldn't tell them what she said. I'd also settle who may hear it early, while she can decide. And if this narrator asks for something to stay between you, I'd switch the recorder off, because a promise to cut it later needs someone who can hear it back.
When she can no longer decide at all, the Society's guidance is strict. It is written for staff, under a law it names, the Mental Capacity Act of 2005: they must act in her best interests, and it limits when a person who can't consent should be involved at all. Your own law may be different. Who may decide for her is a matter for it, and in the United States alone the rules differ from state to state. My own caution is narrower. What was recorded while she could decide stays inside what she agreed to, and if she agreed to nothing, it stays private.
That is what she's told. What she's promised is harder, and the plainest case of it comes from a long way from any kitchen.
In 2001 Boston College began an oral history of the conflict in Northern Ireland, recording former paramilitaries from both sides. Its director, a journalist, has said the narrators could be shot if it got out that they were talking, and its librarian told a court it was doubtful any of them would have talked without an assurance. So each interview was to stay closed until its narrator died. What follows is mostly as an American appeals court later set the facts out.
The narrators signed a form. It said the tapes would stay closed until after the narrator's death. And it went on, "the ultimate power of release shall rest with me."
Before any court comes into it, look at that sentence and ask whose promise it was to keep.
The form was also signed by that librarian. Before the project began, he had told the director that he couldn't guarantee the college would be in a position to refuse a court order. And the director's own contract with the college said that narrators were to be guaranteed their terms to the extent American law allows. Those six words weren't on the form the narrators signed. Neither of the court rulings behind this account says why.
Ten years in, the police in Northern Ireland came asking. They were investigating a killing from 1972: Jean McConville, a widow bringing up ten children, taken away and shot. The college asked an American court to refuse, and the court ordered interviews handed over. The director and one of the interviewers appealed, and lost. The appeals court quoted a principle from the law of evidence: "No pledge of privacy nor oath of secrecy can avail against demand for the truth in a court of justice." It called the assurances on the form misleading, and the situation clearly avoidable.
Interviews went to Belfast, and the police came back for more: another narrator's recordings, and then that interviewer's own.
In 2019 a case built on those recordings came to its hearing in Belfast. The accused was Ivor Bell, who denied any involvement. The judge said the evidence that the voice on them was Bell's was overwhelming, though that would have been for the jury. Then he ruled the recordings out, and the promise was the first of his reasons. The man on them had been guaranteed that whatever he said would stay confidential. That gave him freedom to speak the truth, the judge said, and freedom as well "to lie, to distort, to exaggerate, to blame and to mislead." The second reason was the interviewer, who in the judge's view was no neutral party and had an agenda of his own. Bell was cleared.
So the promise failed twice. It didn't keep the interviews closed, and it made what was said in them worth less as evidence.
It's hard to picture a court order for your grandmother's memories of the farm, and I won't pretend otherwise. But look at what that American court said: a promise of secrecy, however solemn, doesn't by itself hold against a court. I have no source for what your own courts would do, and I wouldn't count on them to differ. So the mistake on that form has a shape that fits a kitchen: it promised what other people would do.
The Oral History Association's principle is that oral historians are careful, at every point, not to make promises they cannot keep. I'd put that as a question. Whose doing does it depend on?
Try it on what you might say in that kitchen, starting with this one: I won't repeat this to anyone. Whose doing? Not yours alone. A court may be able to make you speak, which is what that American court was saying, so this promise carries the six words that were left off that form. And the National Health Service, advising anyone worried that an adult is being abused or neglected, says it's best not to promise that you'll tell nobody.
So the honest version is longer, and its limit about risk is one I'm adding myself. I won't choose to tell anyone, unless I think someone is at risk, or the law makes me. If it came to that, I'd ask you first what you wanted done, and I'd go only to someone who could help. Say it before she tells you.
Now this one: the recorder goes off. That needs nobody but you.
Then Willa Baum's promise, with one addition of mine, which is that she hears it first: it stays on, and when you've heard it, I'll cut that part if you tell me to. Put the question to it yourself. Whose doing does that depend on?
Yours, as far as the law allows, which is the six words again. And it has a deadline, which is the day a copy leaves your hands. If your recorder copies itself anywhere, that day may already have come, so the cut has to reach the copies too. Baum wanted this promise for the record's sake. What it gives the teller is the chance to decide after she's heard herself.
And this one: it's closed until a date you name. That one is yours for as long as you hold the recording, so say who holds it after you.
And the last one: nobody will ever hear this. That depends on everybody, forever, and it's the one sentence on the list you can't say.
Two things are settled, then. She has been told the terms, by you and out loud, and she has been promised only what you yourself will do, with its limits said. What isn't settled is whether any of it gets written down. Somebody has tried the paper, and written down what happened.
Two researchers recorded the life stories of women schoolteachers in Bangalore, some still working in schools they couldn't afford to offend. They began with written agreements, signed by both sides, and found that in their project the forms tended to wear trust away. Two of the teachers who signed weren't sure what rights over their own stories they had signed away. What the researchers promised was privacy as far as possible, false names, and the recorder off on request, and that last promise, they write, won them a great deal of trust.
What am I signing away? That was the teachers' question, and it's a question of law, so one notice, once. What I can tell you about the law is American and some years old, and none of it is legal advice. Your own law is yours to check.
Even there, it isn't settled who owns an interview. The Oral History Association tells narrators that they own the words they speak. The man who wrote a guide to the law for oral historians, a historian who has also sat as a municipal judge, says the interviewer may well own a share. And it may be different again where you live.
On recording itself, some American states make it the law that everyone whose voice is on a recording has agreed. This lecture has asked as much: her yes, and the yes of anyone else on it. Then there are the other people in her stories. She can't agree on her sister's behalf, and that same author warns that what a narrator says about living people can bring claims of defamation, or of exposing private facts, and that putting interviews online multiplies the chances. It's the best reason I have for making online a decision of its own.
Since I can't tell you who owns it, record who decides. The Association would have that in writing, signed, and failing that, said on the recording. For a family I'd turn that around. Have her say it on the recording every time. Write a page when something is to be closed, when the recording is going beyond the family, or when her memory is beginning to fail.
After the session, she hears it first. Play it to her, or give her a copy, before anyone else has one, and ask what she wants cut, what she wants closed and until when, and who may have copies.
If you write the page, it has a name. You heard the word on the Boston College form, the power of release. In this field a release is the page itself: her signed permission for what may be done with a recording. A family's can be short. Which recordings, by date. Who may hear them and have copies, and whether anything goes online. What's closed, and until when. Who holds the recording after you. And whom she'd want asked when she can no longer say. That last line is no power of attorney. It records what she wanted, and who has the power to decide for her is a matter for your own law. You both sign, and each of you keeps a copy.
Suppose you do all of it. She agreed, she heard it first, and nothing was promised that you couldn't deliver. One thing is left that she can object to, and it happened to a granddaughter.
Katherine Borland tells this about her own grandmother, Beatrice Hanson, whom she calls Bea. In the summer of 1944 Bea was thirty-five. She put on a pale gabardine dress with gold buttons, an enormous black hat and gloves to the elbow, and went with her father to the harness races at the Bangor fairgrounds. That day is the story: father and daughter wrangling over who could pick the winner. Some forty years later Borland got Bea's telling of it on tape, and the two of them were delighted to be working together.
Then Borland wrote an essay about the story for a graduate class in folklore, and sent her grandmother a copy. The answer was thirteen pages, typed, single-spaced, by Borland's count. In the essay, as Bea read it, the day at the races had become a woman's struggle for autonomy in a hostile male world. She was not a feminist, she wrote, and never had been, and that reading was entirely her granddaughter's. "The story is no longer my story at all," she wrote, with the word my in capitals. And she asked how far it's permissible to go "and still be honest in respect to the original narrative."
What do you do with thirteen pages?
Borland went back. Over later visits the two of them talked it through and came, she says, to an understanding fuller than either had started with.
Thirty years after that letter, Borland still doubts an interviewer's confidence that she has understood. In its place she'd put a question: "Do we understand each other?" It means going back to the narrator with your reading of her story. "It does not require agreement." And being close may make it harder. Borland cites scholars who suggest that the closer the bond, the harder it is to listen without putting your own meanings in, and she thinks that may have played a part in her own case.
I'd keep three things apart here: her story, the recording, and what you make of it. The story stays hers. The Shoah Foundation takes the copyright in the testimony it films. Even so it tells its narrators that interviewees own their own life story, and the Foundation does not. I take that as a principle, and not as a statement of law. The recording is whatever the two of you agreed. And what you make of it is yours. The Oral History Association lets a reader of an interview reach conclusions the narrator wouldn't, so long as they come from the evidence, and the account stays true to the words and the meanings she gave.
So before you pass on anything you've written or said about her life, show it to her, and ask Borland's question. You needn't change your mind. Keep her answer with it.
And do it while she can answer. When Borland wrote all of this down, her grandparents were, as she put it, no longer around to correct her.
What does she have a right to? By the principles this lecture has followed, which are ethics and not law: to know what it's for and who it's for. To say no: to a question, to the recorder, to the whole thing. To have it cut or closed. And I've added two: to hear it first, and to answer what you make of it.
And what can you honestly promise her? Four things, and each is something you do. That you won't choose to repeat it. That the recorder goes off when she says. That when she's heard it, you'll cut what she tells you to. That what's closed stays closed until her date. Each of them as far as the law allows, which were the words missing from that form and are yours now. The one thing you can't promise is that nobody will ever hear it.
Some of this isn't settled. Who owns a recording. Whether paper builds trust or wears it away. And whether a court would ever reach for a family's recording, on which this course has no source.
All of it assumes a recording worth arguing over, and how you get one good enough to keep is the next lecture.
So, back in the kitchen. She stops, and she says that this next part stays between the two of you. And this is what you can say instead. I can switch it off right now. Or I can leave it on, and when you've heard it, you tell me whether that part is kept, cut, or put away until a date you pick. Either way nobody hears it from me before you do, and I won't choose to repeat it, unless someone's at risk or the law makes me. What I can't tell you is that nobody ever will. Then let her choose.