Media Confidential

Counting the casualties of war

As the death toll in Gaza continues to rise, statistician David Spiegelhalter joins Alan and Lionel to discuss how to better understand how many people are killed in conflict

March 21, 2024
article header image

In any war, counting the number of people killed is challenging. So, too, is understanding how they died. In Gaza, where the still-rising death toll already includes 13,450 children, these figures can be obscured by biases, allegations—and the realities on the ground. In this week’s episode, Lionel Barber and Alan Rusbridger are joined by leading statistician David Spiegelhalter to discuss how to shed light on casualty numbers in a war situation. 

Also this week, George Brock joins Alan and Lionel to discuss a small yet significant development in the future of local news. George is a professor of journalism at City, University of London and has previously worked at the Yorkshire Evening Press, Observer and Times, where he was managing editor and Saturday editor. George explains how the Guildford Dragon has secured charitable status, and whether this could be a possible model for local news across the country.


This transcript has been edited for clarity.


Alan Rusbridger: Hello, welcome back to Media Confidential from Prospect Magazine, our weekly dive into the world of media to find out what’s happening behind the headlines and beyond the clickbait. I’m Alan Rusbridger, and with me in person--

Lionel Barber: It really is me, Lionel Barber, and I am back in the UK.

Alan: In today’s episode, could a lifeline be about to be thrown to the ailing local news sector? We’ll be examining a ruling that allows the Guildford Dragon News to be designated a charity and what this could mean for local news right across the country.

We’ll also be returning to the war in Gaza. The death toll continues to rise, but there’s been some confusion or dispute about the accuracy of the actual numbers of the dead being reported. We’re going to talk to the leading statistician, David Spiegelhalter, on the difficulties of counting the casualties in any war situation but particularly this one. Remember, follow us wherever you get your podcasts to make sure you never miss an episode and you could follow us on X, formerly Twitter, where we are at Mediaconfpod. We’ve lots to cover today, so let’s get going.

Welcome back, Lionel. How does little old England seem after your trip to Texas?

Lionel: A lot colder. I was in California as well, where I managed to pick up half a suntan, as you’ve seen.

Alan: You were cycling in Texas?

Lionel: I was cycling in Texas. I was talking a lot about politics. I was hearing and watching how big business and business people, I think, are circling back behind Trump. It’s a bit worrying, Alan. Obviously, you’ve got months to go before the November election, but you really have a sense in which people are trying to find excuses for why they might actually have to vote for Trump after all.

Alan: Biden’s punchy State of the Union address hasn’t impacted?

Lionel: I think it was massively overplayed in the media, both in the UK but also in the US. Just because he showed some signs of life and punchiness, everybody assumed that this was the star. I think it so much depends on the economy, whether the interest rates look likely to be cut. That may help, but it might come too late, like in 1992 with George H.W. Bush.

Alan: Meanwhile, back in Blighty, there was one bit of unsurprising news, which was that Robbie Gibb was reappointed as a BBC Governor. There was a half-hearted attempt in the Sunday Times by BBC insiders to push back on our reporting and our discussions, saying that actually, no one takes him very seriously and he’s a bit of a laughing stock internally. Anyway, he’s not so much a laughing stock that they haven’t reappointed him for four years through an opaque process that I don’t understand.

Maybe more interestingly, Ofcom has--

Lionel: They’ve been listening to you, Alan.

Alan: They have done something. They have found that five programmes have breached impartiality rules. I suppose we should be thankful for small mercies. It’s a pretty limited finding, and they’ve said they’re jolly cross and if they do it again, they might be even crosser and they might even think of getting a slipper out and wrapping them across the knuckles. It’s a little weak, I think, and it ignores the systemic problem that’s going on at GB News, but it’s something.

Lionel: I think the door’s opened and the issue now is whether GB News takes any note at all, whether behaviour, conduct, the way they are getting around these impartiality rules, due impartiality, I should say.

Alan: Due impartiality.

Lionel: Or whether they just stick two fingers, to use a technical term, at Ofcom. I think Ofcom discovered a little bit of courage, and I’d like to think this podcast and your steady interventions, Alan, will have influenced the outcome.

Alan: Maybe a tiny bit. The other thing that I did this week was to go to a breakfast yesterday in honour of Harry Evans, who died five years ago, and all credit to his widow, Tina Brown, who’s trying to use his legacy to build something. She’s got a movement called the Truth Tellers. It was an interesting intervention from James Harding, the former editor of The Times and former head of news at BBC, who raised an article in the Sunday Telegraph by the former Attorney General Michael Ellis, who’s a figure I was once interrogated by in the Home Affairs Select Committee.

Lionel: Over the Snowden Affair.

Alan: Over the Snowden Affair. I believe he was once a star of the Northampton Bar before becoming Attorney General, and he wrote a coruscating piece in the Sunday Telegraph saying that the BBC was institutionally anti-Semitic.

Lionel: That’s a big charge.

Alan: James Harding, in front of a very distinguished audience of journalists yesterday, just let rip at him and he described this as absolutely disgraceful, very dangerous thing for journalists and writers in our own country. James said, I say this as a Jewish person who spent five very happy and proud years looking at the BBC and I thought that was good of James because it’s such a easy smear, lazy smear to throw around and you would hope that someone who’d been Attorney General would have a bit more judiciousness before throwing around labels like that.

Lionel: Institutionally anti-Semitic, it’s a huge charge because it suggests that the culture is completely warped and it’s top to bottom. You remember the charge against the Met Police, it was institutionally racist. That has echoes. Of course, if you’re talking about a news organisation, which has a duty for duty of impartiality, it’s very difficult. I see Tim Davey was talking about how a polarised Britain is making it very, very difficult for the BBC to operate. He’s still keeping his kid gloves on regarding politicians. I suppose he has to as Director General, but having James Harding step in is useful.


Alan: First, we’re going to return to the war in Gaza and the rising death toll as reported by Hamas. Is the figure being quoted by sources accurate? Jake Wallace-Symons, editor of the Jewish Chronicle, posted on X that he believes, and I’m quoting here, “Hamas may have pulled off one of the biggest propaganda coups of modern times. The figures, repeated by everyone from the White House to the BBC, are freighted with familiarity. 30,000 dead in Gaza, 70% of whom are women and children. Yet it now seems overwhelmingly likely that these statistics are fabricated.”

He also wrote a piece in the Daily Telegraph, and Jake, who has, I should say, appeared on Media Confidential, quotes the work of a certain Professor Abraham Weiner, a data scientist at the University of Pennsylvania. Weiner has conducted a thorough analysis and claims to have found that Hamas’ official civilian death toll was statistically impossible. This is according to Weiner, and again we’re quoting, “Most likely the Hamas ministry settled on a daily total arbitrarily. We know this because the daily totals increased too consistently to be real. They then assigned about 70% of the total of women and children, splitting that amount randomly from day-to-day. They then infilled the number of men as set by the predetermined total. This explains all the data observed.”

Lionel: I’ve not heard about Professor Weiner, who comes from UPenn. It’s a respectable university, certainly. The big question, I guess, is how do you actually go about counting the casualties of this, or indeed any conflict? With so many deaths, are authorities expected to count every single one? Is it inevitable that one or both sides will massage the figures to suit their agenda? You’d certainly need an incredible maths ability to be able to understand both the claim made and the rebuttal.

Alan: You certainly would. I’ve looked at some of the argument that’s raging around this. As a humanities graduate, I really can’t make head or tail of it. Somebody who can, and joining us now to shed light on where you begin with statistics in a war situation, is the leading statistician, David Spiegelhalter, from the University of Cambridge. Now, we won’t be talking about the specifics in this current case. David is here to explain what you would need to do to assess the trustworthiness of claims based on the statistics he’s seen.

David, welcome. Thank you for joining us. I’m afraid you’re talking to two humanities graduates. We were rather embarrassingly looking at some of the argument that’s raging around these articles. We realised we couldn’t even pronounce them. There are funny things with Xs with a number below it. We can’t understand a lot of this stuff. Perhaps you’ve read the article and some of the dispute raging around it. Can you tell us what’s at stake here?

David Spiegelhalter: There’s two things. We can look at the specific example about Gaza casualties. We can also discuss, which I prefer to do in fact, more generally. I’m not an expert on the Gaza casualties, I’m not going to express an opinion about those. Basically, the crucial thing is that data does not speak for itself. There’s always a layer of interpretation that’s going to be put upon that, and here is some data that’s been issued and conflicting interpretations are going to be put upon it.

As a critical interpretation that, for example, took a fairly short section of data and says that it’s too smooth, that the number of deaths recorded each day last October and November were just too even to be reasonable, and that the split between men and women, and women and children is not what you’d expect from a generally random sample.

On the other hand, people have said, a number of academic papers and things, not looking at that specific data, but have said that actually, the Ministry of Health in Gaza has in the past had quite a good reputation in terms of their reporting on data, that there’s a good death registration system in Gaza, and that their stats in the past have been accepted rather broadly and have fitted with other sources of information, and that the counts over a certain period also fitted with really verified deaths of UN personnel, for example, at rather similar rate per thousand people. You’ve got exactly the same numbers being put under two different perspectives.

Alan: From a media point of view, your average reporter in a newsroom, you’ve got PhD-level arguments about sets of data where even people who are highly schooled in handling this kind of data can’t agree. How on earth is an average reporter supposed to approach this kind of data?

David: Very challenging. We just have to go back to the pandemic when discussing things like COVID deaths or at least four different definitions, being passed around with a lot of dispute about what the data means. That dispute is still going on about excess deaths, for example, which I happen to work on. How do you--

Lionel: Excess deaths meaning?

David: Well, exactly what does it mean because it’s not a well-defined term. You can’t count excess deaths. It has to be excess over something else. A counterfactual of what deaths you would’ve expected in a different circumstance, and what circumstance you use for that. Is it pre-pandemic? Is it what we would’ve expect given all the data so far ignoring perhaps the worst years, is a matter of judgment and choice by whoever is releasing the data, and some people don’t like those judgments and choices.

I always say that data doesn’t speak for itself. We imbue it with meaning and there’s always a role of judgment both in terms of just what’s collected, and how it’s reported, and then when you actually have to be a consumer of this, either as a journalist or a member of the public, and it’s difficult.

Lionel: Now, David I’m not going to ask you to be a journalist, but are there any tips that you would offer when assessing something like this, casualties, deaths, in a conflict presumably at least more than one source? What else?

David: Yes. It’s not just in debts and conflicts, it’s almost any number. You just look at a claim made about the harms of ultra-processed food, the harms-- all the claims being made all the time based on statistical analysis. Journalists have a challenge, but the members of the public have got a challenge as well because they’re being bombarded with these numbers, whether it’s on social media or through more mainstream media.

People have written various checklists and things like that, and I think these are ones you would do yourself. I, for a start, my main concern is trustworthiness. Is this worthy of my trust, but there’s different aspects in which you need to investigate that. I put number one, even before I look at the number, I look at the source. Which is I think would be a journalistic approach as well. Who is telling me this? What are they trying to make me feel? Are they trying to frighten me, reassure me, shock me, upset me? In other words, what’s in it for them?

Then even more difficult is, why are they telling me this, and even more difficult, what am I not being told? What are the gaps? For example, in the critical report on the Gaza deaths, he’s got a plot saying this is too smooth, but he’s only taken one section of the data. He’s selected the smooth part of the data.

Alan: Over a time period.

David: Yes, over a time period. If you look back a bit, it’s very jumpy, but you wouldn’t know that from what you’re looking at.

Lionel: Snapshots are dangerous.

David: Snapshots are dangerous, and the problem is, just by looking at a set of data or a claim, you cannot tell how much it’s been selected. Again its multiple sources you have to look at. What I’ve heard described is you have to search sideways rather than downwards. If you just look at a particular claim, you go down and down the rabbit hole of how, why they’re basing this claim. You’re never thinking about, well, is the whole basis of this reasonable? You have to look sideways [unintelligible 00:15:32] [crossing]

Alan: What happened before and what happened afterwards.

David: Well, what other people are saying about this claim, what other people are saying about this very data set. You have to look sideways. You have to look at multiple sources about the same claim, is the number one thing because you cannot-- sometimes claims just give themselves away just from themselves. You can tell, this is really roping, just by the way they’ve done it. [unintelligible 00:15:54] they’ve selected data, you cannot tell that without looking at some external source.

Without even you look at the number particularly, it’s purely looking at what are the motivations behind the communicate tool. Unfortunately, I put that now number one of my trustworthy. Then if you get past that, you look at--

Alan: Let me just [unintelligible 00:16:15] because again, I’m not in your field.

David: This is standard stuff [unintelligible 00:16:21] journalists would do all the time

Interviewer: Have you heard Abraham Weiner?

David: No, I haven’t checked him at all. I haven’t done my research.

Alan: In order to decide whether he was trustworthy, would I look-- is it the fact that he’s University of Penn [unintelligible 00:16:33]

David: Doesn’t necessarily mean very much. There’s just a lot of very opinionated people working in universities. Oh, I would google him. I haven’t done it in fact, but I’d Google him, I’d check his social media accounts, and so on. That basic investigatory stuff, which really, it seems absurd for me to have to do as a statistician. It’s what I do all the time whenever I read anything. I read a scientific paper, I look at who it comes from, what have they said in the past, have they got a strong agenda?

It doesn’t necessarily mean they’re wrong, it’s just to be warned about that. The next thing is I think you do have to look at the claim, the actual numbers. I always say you got to sniff them. Is this a smelly number or not? Is it a sweet smelly number?

Lionel: I did the same with my sources, but actually, we won’t go there, David.

David: [laughs] Exactly, no. You’ve got to sniff the number. Rather absurdly, one of the first things is, what are you actually talking about? You have to look at the definitions. What are they actually counting? I’m a big fan of More or Less, the Radio 4 programme. I’m on it quite a lot. So many of their numbers they take apart, are just absurd because it’s not what you think they are. They’re what they’re actually counting, it’s just the definitions aren’t right.

That’s one thing, but then you have to look at where did this data come from? How is it collected? Who is responsible for it? Can we trust the actual counts that are being made once you’ve decided what it is? That’s really important to know the provenance of the data.

Lionel: Then there is that intangible which you alluded to, which is gut feeling. How much is gut feeling?

David: I’ve been doing this for 40, 50 years.

Lionel: You don’t look at David [unintelligible 00:18:11]

David: No, I know. I [unintelligible 00:18:12] certainly 45 years of tearing apart claims. I’ve referred zillions of medical papers, something I think now should be passed over to AI. I really do think AI should be doing this, taking apart medical claims. I’ve done so much, it made me a bit cynical, skeptical I suppose. I always say, statisticians are often called evidence policemen, that they’re in there when a claim is based on statistical evidence. We go in there and my immediate reaction is, I don’t believe it. I’ve got my groucho principle for numbers in the news. The very fact that they’re in the news is reason to disbelieve them.

The fact that I’m hearing this is reason to disbelieve it because it wouldn’t be there unless it was surprising. A claim that was in the newspapers yesterday, restricted diet, intermittent fasting, only eating over eight hours. 91% increased risk of cardiac death is the claim, it was in there. Now, it got coverage. Now I wrote with the science media saying this is awful, this is terrible. We can’t conclude this. It should never have been press-released. We’ve got no real data. It could have been all sorts of reasons.

This is just based on years ago people filling up a food questionnaire about when they had their meals for two days, maybe they just forgot some meals. How were these two days chosen? We don’t know anything. It was an abstract for a conference, front page of The Times.

Alan: Looking at this data and you’ve described what you would do with this data. Is this a PhD level task? Is it a professor task? Do you have to have professor knowledge or is it if you’ve got an A level in economics, could you--

David: Absolutely, anyone could do this. It doesn’t require technical skill whatsoever. It’s much more it requires I think quite a lot of experience for realising what people can do with data. Because even the statistical claims, there’s a simple correlation, that’s A level [unintelligible 00:20:05] The claims about the data being not having a particular pattern or A level. It doesn’t require technical skills at all.

Lionel: On the other hand, you’ve got, I assume, to be reasonably eminent professors completely at odds over what the data is telling us.

David: Yes, but it’s not to do with any technical analysis. Sometimes it is, but almost always, the conflict has been between whether you just trust the numbers, whether you can believe them or not most of the discussions about the reliability of the source that they’re saying in this case. Also, you could be arguing about even if the source was reliable or unreliable, whatever the reliability of the source, whether the numbers themselves are accurate or whether they’re just missing cases, which is definitely the case.

There’d be missing cases, but also, they’re counting all casualties rather than just military casualties, either counting all deaths rather than military. There’s all sorts of ways in which the numbers themselves need checking.

The other thing, which this particular critic did, was anomalies in the data, just as a set of numbers almost. Are there anomalies in the data? Too smooth, they don’t fit a pattern that you’d expect by round variability.

Alan: Looking at the static around this, and you’ve read your way into this a bit, are there people arguing on both sides, or is it a pile-on from one side against the other side?

David: Oh, I’ve seen stuff on both sides because there are academic papers written last year, London School of Hygiene, Johns Hopkins, saying these were fairly reliable figures, and there’s a lot of critics.

Alan: What’s the responsible way for the media then to handle this? You’ve done enough media to know the constraints of reporting. You’re Jeremy Bowen, you’re standing in the middle, you can’t get into Gaza but on the edge of Gaza, and you’re quoting the Hamas figures. Is it responsible to say, though I should qualify that by saying that these have been contested by some and verified by others?

David: I think the media are clearly obviously getting around that, partly by always saying it’s a Hamas-led Ministry of Health. Always saying that, putting that as a preface before mentioning the Ministry of Health.

Alan:: Do you think that’s enough?

David: I think that’s probably quite good. I think it certainly gives-- immediately, it warns any reader that this is something to be taken carefully. I think that works pretty well in fact.

Lionel: It seems to me if I was editing the BBC or in my old job, you always got to be wary of easy numbers which attract the headline. 300,000, 200,000 rounding errors. That’s one thought, but the second, David, is, do you think that you can ever have precise numbers when it comes to a war?

David: No. That’s why when you see 30,672, they think, what this nonsense? They really shouldn’t give those because giving that count like that gives it a spurious impression of accuracy, which is completely inappropriate. Just in the way the number is communicated, one can give a feel for the fact that it’s not a reliable count. I’d say exactly the same with COVID. I ranted on against these daily numbers. When some minister would get up and say, “Oh, 37,622 people have died from COVID.” What a lot of nonsense, absolute nonsense. He doesn’t know that. We don’t know. What do you mean by dying from COVID? He is just spurious, what I call number theater, I would say.

Lionel: Even worse, the likely number of deaths, if certain preventive measures, lockdown measures were not introduced. Wasn’t that a bit of a numbers [unintelligible 00:24:04]

David: Oh, yes. That was a half million, which is [crosstalk] if we all just sat around folding our arms and twiddling our thumbs and doing sudokus while the epidemic raised through the country, half a million people would’ve died, but that’s not even a reasonable worst-case scenario. That’s a lunatic worst-case scenario.

Alan: In general, David, do you think the media is getting better at this?

David: I think they are. I think COVID, really, it was a step upwards in the quality of reporting of data in the media. The use of graphs, the critiquing of data, whether it’s the models, the claims, the vaccines, or something like that. No, I think it has got a lot better. I’ve been very encouraged if it’s done by people who actually know what they’re talking about. The disaster is when it’s handed to-- actually, some of the political reporters are not good. They get off on stuff without understanding it. The general journalists can be hopeless.

I give talks on this. The quote I gave to the Evening Standard, that I was always saying, “Oh, COVID deaths soar to 200. COVID deaths soar.” I got three COVID deaths soar to this thing. The point about those, I said, they only puts those anything in common between those, they’re all on Tuesdays. The deaths reported on Tuesdays were always hugely high because they soaked up the whole weekend because they weren’t reported over the weekend. They were all on Tuesdays. Everyone who knew anything about it had learnt to ignore Tuesday’s and Wednesday’s data.

Lionel: I think that’s true. Also, Evening Standard sales were always generally low on Tuesday. [laughter] Well, David, you’ve been really fantastic at explaining this. I just think you’re wasted as a statistician. You clearly should have been a top reporter or even an editor.

David: Look. Certainly not. No. I’ve really enjoyed working with journalists throughout COVID. I really worked with almost all journalists. Not everyone, but nearly all, because raising the level of discussion right across the media about stats and numbers, it’s in everyone’s benefit. They should not just be used for weapons for advocacies on one side or another.

Alan: Thank you for joining us today, David.

David: Thank you.


Lionel: Well, I thought that was a fascinating discussion. I also thought that David offered some really useful practical tips to both editors and reporters in dealing with casualties in a conflict, how you assess sources in many ways. I was being serious. He was adopting good journalistic practice.

Alan: He was. He was telling us how to do it. He was careful not to reach his own conclusion because we talked to him afterwards. He said it would take quite a lot of work to work out the full context of the numbers and to try and understand them, and to place in the context of what we know about how these numbers are being gathered, the record of the Gazan Health Ministry in the past, and so forth.

The beginning of this was the claim that this was the biggest propaganda coup of modern times. I don’t think we’ve established that by any manner of means from the basis of what David’s told us about the methodology.

Lionel: He was careful not to go into the details, but there was one I thought big, what I used to call magic moment when he talked about the dangers of a snapshot, and in the [unintelligible 00:27:37] analysis, this was a particular--

Alan: It’s a limited time period.

Lionel: Limited. Exactly.

Alan: The suggestion was, we don’t know. I think he said before and afterwards, it wasn’t so smooth. That’s why there’s claim that these numbers are unbelievable because they’re so smooth. I think David was saying, if you look on either side of that, that’s not true. We don’t know why this particular timeframe was selected.

Lionel: He was also talking again afterwards about the dangers, or not the dangers, but more if you see something that looks anomalous. If there’s a particular spike in it, does that correlate with what was happening on the ground? These are, again, what any good reporter-- obviously tremendously difficult circumstances in Gaza because you’re not allowed in, but I thought you made the good point that if you’re Jeremy Bowen sitting outside, what do you say?

Alan: I think it shows how you can’t win, because if Jeremy Bowen in future just says these are Gazan Health Ministry figures, he always puts the writer in saying-

Lionel: Hamas controlled.

Alan: -Hamas controlled. If he now doesn’t add a writer saying, these are contested figures, blah, blah, blah. In journalism, you’re always tight for time and space. I suspect the reason for discussing it today is this will become another big debating point about the figure about the war. It’s just good to have unpacked it a bit.

Lionel: I think if you can spare the resource, it really is worth having a data unit in a news organisation. If you can’t do that, at least give some of your reporters some basic training in understanding data.

Alan: Yes. I was disappointed actually. I didn’t see an analysis of it because it’s a very inflammatory claim to claim these because they’re all nonsense and its just propaganda. It’s what newspapers should do really, isn’t it, to say, well, look, here are two sides, where does the truth lie?

Lionel: Dig deep. This is Media Confidential. Coming up, we’re joined by George Brock. George is a professor of journalism at City University of London. He’s had a glittering career to date. He worked at the Yorkshire Evening Press, The Observer, and The Times, where he was managing editor and Saturday editor. Here’s the deal, he was Brussels correspondent back in the early ‘90s when a certain Lionel Barber was the Bureau Chief for the Financial Times.

Alan: You were all colleagues, and slightly eclipsed by, if I’m not being unfair, one Boris Johnson.

Lionel: If you describe him as a decent, accurate, hard-working reporter.

Alan: I didn’t say that. I said he eclipsed you.

Lionel: As George and myself.

Alan: You were a star in the firmament. You were little moons.

Lionel: I used to scoop him regularly, actually, Alan. We’ll come to that in another podcast.

Alan: You used to make the stuff up. In this week’s Prospect podcast, Ellen Halliday is joined by Diane Sugic, the former director of the Design Museum, to discuss California Forever, an amazing plan to build an entirely new, affordable, community-focused city in Solano County. How do the locals feel about it?

?Speaker 4: A couple of local congressmen, both Democrats, have denounced California Forever as playing hardball, mob-like tactics, intimidation. There have been lawsuits brought against some of the landowners for trying to rig the market by asking for inflated prices. People had their leases for farming on some of this land, terminated land that California Forever has already bought. They’ve thrown people off the land for criticising them. They’ve not been playing nice.

Alan: Be sure to follow the Prospect podcast wherever you get your podcasts. Also, you can enjoy an upgraded digital experience with the new Prospect app. Stay informed and engaged with our independent journalism at your fingertips. Read or listen anywhere at any time. The app not only gives you access to our articles, but also includes audio narration, plus podcasts, like the one you’re listening to right now. You can subscribe to Prospect and get instant access to the new app, which is available on iOS and Android.


Now, we’ve touched on this story in previous shows. There’s a crisis in regional media, we know that. Local newspapers are being swallowed up by three big players; Reach, Newsquest, and National World. Titles are being consolidated. Local knowledge and expertise is being lost as journalist numbers are in freefall.

Lionel: Press Gazette study in 2022 looked at the financial numbers across the three dominant media organisations and compared them to the nine companies that were operating back in 2007. The figures do not make good reading. In 2007, there were just under 9,000 journalists working across local titles. By 2022, Reach, Newsquest, and National World employed under 3,000.

Alan: Staggering figures.

Lionel: Advertising revenue also fell from £2.4 billion to £590 million. There were further job cuts in 2023. One of the questions we need to ask is, how can these newspapers and websites function?

Alan: That’s what we’re going to talk about today because there’s a tiny glimmer of hope, which is whether the future of local news could lie in their ability to operate as charities. The first major step in this direction was taken last month when the Guildford Dragon News, a Surrey-based website, was officially registered as a charity. We’re joined now by George Brock to guide us through these complex waters.

George, I think you’ve been involved for some years in this project to try and get charitable status for forms of news. There’s been a breakthrough. Can you tell us what that breakthrough is?

George Brock: Yes. Some years back, as you say, a group of editors and journalists and academics and lawyers thought that it was particularly difficult for local and community news organisations to get charitable registration, and that would be suitable for some of them, if not for all of them. There turned out to be an awful lot of resistance from the Charity Commission to doing this. It’s not an established charitable purpose in law.

Eventually, after a great deal of research and argument and debate and so on, we have managed this February, or we have been part of registering a local publication called the Guildford Dragon News as a charity. The key thing is that when registering them, the Charity Commission have acknowledged that what they are doing is public interest local news.

That’s the first time, that’s only the second time they have ever registered as a charity a journalism provider, as opposed to organisations which discuss journalism. It’s the first time that they have registered as a charity a local site and said what it’s doing is local public interest news. While there are still a few stages to go, we thought that was a pretty big breakthrough in the world of this debate.

Lionel: Before we look at the implications, I would like to hear a little bit more about the Guildford Dragon News. When was it established? How many people does it employ? What kind of stories is it following?

George: It was established about 10 or a dozen years ago, and significantly, were involved people who had been made redundant from local papers in that area. People looked at new local news being flung around on Facebook and Nextdoor and online sites of that kind, and thought, well, we can’t rely on the existing legacy newspapers that are around here, they’re edited from too far away, they don’t have enough reporters, they’ve lost interest in our area, but we can set something up ourselves.

They sat down and tried to work out how to do it. They don’t involve full-time more than about three people, I don’t think, and I’m pretty sure they’re all retired at the Guildford Dragon News. They’ve got all sorts of part-timers. The guy who looks after the business insofar as they’ve got a business is also the beekeeping correspondent, if I remember right. They struggle on, but they are very, very dependent on the charity of their readers, occasionally to drop them a little bit of money. They don’t charge for what they do. They have one or two philanthropic funds, so they have the income that is the usual mixture for these online local start-ups.

Lionel: There are two benefits that I’ve seen you talk about, George. One is that there are tax breaks associated with this money, so that if somebody gives money, they can leverage that. The second one is you think there’s a reputational gain to the organisation. Maybe not just a reputational gain, but they have to satisfy the charity commissioners that they’re working responsibly and doing what they said they would do. Can you talk a bit about those two benefits?

George: Yes, the tax break is simply a 20% tax break on donations, which if you’re asking your readers for £5 a month or £10 occasionally or whatever, still matters. Also it’s significant because quite a lot of philanthropic funders, and there aren’t very many involved in local news, but there are a few, they often will only give money to registered charities. That’s a double break if you like.

The reputational gain is that if you are trying to be careful to say, we are being objective and fair and we’re not pushing any particular line or any particular political party or whatever, then to be able to say, and we are regulated along those lines by quite possibly more than one external regulator, the Guildford Dragon News is also regulated by Impress, one of the post-Leveson regulators, but there is also the Charity Commission who have laid down that this is journalism done to the highest standards in the public interest.

I’ve no doubt, because this is a breakthrough, that there will be an awful lot of people examining exactly what the Guildford Dragon News does, and making sure that it stays at the standard that the Charity Commission want.

Lionel: George, you mentioned that there are still one or two hurdles to go through. What are they?

George: The hurdles are, well firstly, that an awful lot of people are going to examine this with a microscope. There will probably be debates and disputes about who is entitled to charity status. Anybody can look at the charitable objects, the rules which the charity commission have laid down for the Guildford Dragon News, and one of the things that has been a very important part of the breakthrough is that the Charity Commission have said, “We don’t require every single thing that appears in the Guildford Dragon News to be highly serious public interest news. You can put entertaining things in if you want to, if you need to keep up the attention given to your publication and it doesn’t overwhelm the whole thing. It remains a minority of what they do.”

Now, almost all of the subtle distinctions that are in the Charity Commission’s public declaration, anyone can look it up, are bound to be debated in the future. Of course, charity registration doesn’t guarantee that the Guildford Dragon News’ income is going to rocket upwards. It certainly isn’t. I hope it rises gently because it’s able to make known to its readers and potential users that it will do that. Charity status wouldn’t suit everyone. The Guildford Dragon News’ approach to news seems to me to be very straight and very serious, but that doesn’t apply to every community news site in the country. They vary enormously.

Alan: You could still have a mixture of news and comment, pungent comment?

George: Oh, absolutely. There’s quite a lot of misunderstanding about what the Charity Commission thinks is fairness and objectivity and lack of political bias. They do not rule out strong comment or strongly worded comment or strong opinion. What they don’t like is something that is a consistent pursuit of the program or manifesto of a single political organisation.

The Guildford Dragon News carries perfectly strongly worded editorials about local politics and things that are going right and wrong in Guildford, and that hasn’t prevented them getting registered as a charity, and it won’t stop them doing all that kind of thing in the future.

Alan: If I was running Reach and I thought, I know what, I’m going to turn the Manchester Evening News into a charity, is that imaginable?

George: It’s imaginable, but it’s not very likely. Of course, it’s theoretically possible. If they set out to be as serious and dedicated to their community, as I think you can say the Guildford publication is, and others too that I hope will now apply, then they can demonstrate all of the things that they need to demonstrate, which is public interest news, local public interest news. If they were Reach, then they would have one massive advantage that the Guildford Dragon News didn’t have, which is they would have the money to employ the specialist lawyers who you need to shepherd one of these applications through the Charity Commission’s fairly obstacle-strewn and slow and difficult process.

Alan: Just so I understand, why would you say the Manchester Evening News was not dedicated to community news in the sense that the Guildford Dragon News is? I’m sure Reach’s lawyers would make a very persuasive argument that they are.

George: They would look at the degree. For a start, they would look at the degree of engagement and how reporters were used and spread around and was it closely enough connected to the community to really deliver the majority of its material as public interest news. I’m not saying the Manchester Evening News couldn’t show that, but obviously, that would be a radical change of their business model. I think in practice, but I’m guessing here, it’s unlikely that publications as large as the Manchester Evening News are going to try it, but there’s nothing in theory to stop them trying.

Alan: Your old paper, The Times of London, can you imagine them wanting to register as a charity?

George: I think there is a very strong stress in what’s called the statement of charitable objects that gets published when a charity gets registered. The one about the Guildford Dragon News makes it absolutely clear in every line that this is about the benefit to a local community. I think you’d have a very much harder time taking any national publication claiming that it would have the same demonstrable benefit.

I’m not a charity lawyer. Who knows? There might be publications looking at this news and thinking, oh, there are all sorts of ways we could exploit this. My feeling is that the Charity Commission have probably thought about that possibility and done their best to limit it to a local scope.

Alan: Just to pursue that thought though, George, I know my old paper, The Guardian, in America has registered as a B Corp. You get a special tax break and you’ve got papers like the Texas Tribune that are philanthropic models. Can you see the drift in this direction that papers that can make out that they are there for the public benefit, in the public interest, should get some kind of relief?

George: I can easily see things going in that direction. I think it would take quite a long time to happen in practice. The way charity law is written in the UK would make that quite a difficult, if not impossible, development to look at. However, if you had a government which said, we not only have to worry about local journalism where you can argue the crisis, the business model crisis is at its worst, but not only do we have to worry about local journalism, but we’ve got to worry about publications covering a much wider scale and scope and we’re going to do a whole series of things.

One of the things we’re going to do is make it much easier for them to receive, for example, donations and get a tax break and so on. Whether you can adapt the current charity law, but it’s certainly not inconceivable and I think if the business model crisis continues as bad as it is, it may eventually meet a government sufficiently sympathetic and serious to try and do something about it. Like I say, I don’t think it’s right around the corner.

Alan: You’ve looked at the American model, George, and how would you describe the difference between the American philanthropic model and the state of affairs in the UK?

George: There are two really big differences between the US and the UK. In the US, there is far more charitable money available. That’s a function of a lot of deep interest in journalism, but it’s also a function just of the amount of money being generated in American business, I suppose. The second difference is that the law is just simply much more generously drawn in the US. It just makes start, in general, leaving not particularly confined to journalism, start-up, the law and culture of start-ups is just much more enabling in America than it is in the UK.

Alan: If you were advising the labour government, I know you ideally wanted to see something in the media bill and I think it’s too late for that. What would be the most imaginative thing that a Labour government could do to create good conditions for these start-ups in local communities?

George: The simplest thing would be to legislate, if possible, for journalism to be established as the 14th formally written-in charitable purpose in the law, but if legislative change was too much, if a government said, we note with pleasure that the Charity Commission has registered the Guildford Dragon news and we strongly encourage it to think sympathetically and constructively and imaginatively about these kinds of applications, that in itself would do quite a lot.

Charity law is written so that it’s flexible and can be adjusted. You don’t absolutely have to have a purpose written in which covers what you’re doing, and the charity commissioners are explicitly allowed to adapt for new things that might be important as charitable purposes, and that’s the flexibility that they’ve used in allowing the breakthrough in Guildford.

Alan: George, thanks so much. It’s a fascinating breakthrough and hopefully a significant one. Let’s hope there are many more dragons to--

Lionel: Not to slay. I knew you were going to mix your metaphors there.

Alan: I was searching for my metaphor, thank you. I was thinking of fiery dragons burning their way through loopholes, but that would be a mixed metaphor, so preposterous, and I’m glad you rescued me from that. George, thank you very much.

Lionel: Thank you, George.

George: My pleasure.


Lionel: That was a heartening story, result of several years of campaign and it’s a burst of light in a pretty grim landscape for local news.

Alan: I think it’s a great story. I know the Bureau of Investigatory Journalism has tried three times and I think they’ve succeeded on the third time in getting partial recognition for the work they do. I think it’s a grown-up decision by the Charity Commission recognising that in a world, I think, the startling figure that there are fewer local newspapers now than in the 18th century, that it’s bad for democracy to have news deserts and the vacuum that you get in a news desert where the bad sources of information come in. I can see, I think it is evidently a charitable good to have people who are trying to produce good, clean, reliable information.

Lionel: Yes, I thought everything I agreed with, I wasn’t quite sure whether it would be helpful for the government to make a big public statement. Might be just worth a little bit of a quiet word in the air and gentle encouragement because-- well, just I think then we don’t want people jumping up and down saying, a future labour government or Lib Dem or whatever are actively interfering in the news space. It’s just could be easily distorted that the sentiment that you articulate is absolutely correct.

There has been clearly a market failure in the local news area because as a result of technological change, as a result of the disappearance of advertising, these local news organisations just don’t have the strength to survive.

Alan: What I like very much is these new startups who are very focused on the bread-and-butter of councils, courts, how public authorities are spending our money and so on and so forth.

Lionel: Which is the democracy point [unintelligible 00:51:07]

Alan: Yes. What happened in Australia, when Murdoch effectively got Facebook in an armlock and said, “Give us your money,” Facebook for [unintelligible 00:51:20] themselves handed over a large sum of money to the existing players. You just think, well, okay, [chuckles] handing all your money over to Rupert Murdoch perpetuates a model of journalism which some of it is excellent, some of it not so much. I like these new, really quite lean dedicated startups. We interviewed Yoshi Herman a month or so ago, and they seem to me a pointer to the future.

Lionel: Indeed. I noticed as well that a number of the journalists in the Guilford area were retired, working hard, so I think--

Alan: [inaudible 00:52:07]

Lionel: Yes. I think so.

Alan: Lionel, that’s it for this week. I think I know where you’re going to be next week, you’re going to be in London. Is that right?

Lionel: Definitely here in Blighty.

Alan: For at least one more week.

Lionel: Then off to Japan.

Alan: I knew it couldn’t last.

Lionel: Let’s talk about next week. We’re recording a special episode in front of a live audience. Hold your seats.

Alan: Hold your seats here in Prospect Towers. Well, we look forward to that. Meanwhile, today, thank you for listening to Media Confidential brought to you by Prospect Magazine and Fresh Air. The producer is Martin Poyntz-Roberts.

Lionel: You can send any questions or comments to or get in touch on X, formerly Twitter. We are @mediaconfpod.

Alan: Remember to listen and follow Media Confidential wherever you get your podcasts, and join us next week for more invaluable analysis.

[00:53:13] [END OF AUDIO]