Flower in the River: A Family Tale Finally Told

When Research Starts Talking Back

• Natalie Zett • Season 4 • Episode 148

Share episode

Send us a text

When Research Starts Talking Back

What happens when your research doesn’t just sit there quietly… but starts nudging you, whispering, insisting you dig deeper?

In this episode, I try something a little different. After sharing my 2025 retrospective, The Search Goes On — Coincidence. Clarity. Resolve., I handed that episode to Google’s NotebookLM—an AI tool many genealogists are exploring—and let it analyze the work.

The result? Two AI research companions, Eva and Max (NotebookLM’s AI hosts), listen to my last episode and talk back—analyzing the research, the discoveries, and the questions it raises. 

What surprised me was where they lingered: accountability, documentation standards, and how historical tragedies are sometimes framed and fixed in place.

It’s thoughtful. It’s a little strange. And it’s unexpectedly illuminating to hear your own work reflected back by an algorithm.

You’ll also hear me reflect on:

  • provenance—and what gets lost without it
  • pattern recognition and persistence
  • the messy beauty of family history
  • why history is never really “finished”—and why the inquiry must continue

Coincidence, clarity, and resolve all make return appearances.

What emerges is a case for clear sourcing, shared definitions of casualty, and open access.

Resources: 

Natalie Zett:

Hello, I'm Natalie Zett, and welcome to Flower in the River. This podcast, inspired by my book of the same name, explores the 1915 Eastland disaster in Chicago and its enduring impact, particularly on my family's history. We'll explore the intertwining narratives of others impacted by this tragedy as well. And we'll dive into writing and genealogy and uncover the surprising supernatural elements that surface in family history research. Come along with me on this journey of discovery. Hey, this is Natalie, and welcome to episode 148 of Flower in the River Podcast. I hope you're doing well. I've got something different on the docket for you today. After I published the last episode, and that would be episode 147, The Search Goes On. Coincidence, clarity, and resolve. This was the retrospective of my podcast highlights for 2025. This episode struck a chord in a lot of people, and I did not expect that. It might have been because I mentioned that I was interviewed by two of the genealogy giants. That would be Krista Cowan at the beginning of 2025 and Lisa Louise Cook at the end of 2025. Stories that live in us and genealogy gems. That probably didn't hurt. But we need to move on. When it comes to the history of the people of the Eastland disaster and the way that narrative has been presented in the last few years, well, I made a discovery that changes all of that. The discovery of the actual documented record of a person who was left out of the initial estimates that George Hilton created, and those estimates were in turn used by others, and the others who use the estimates either didn't think to check George's work or just didn't check George's work. Regardless, that number that has been repeated over and over again is incorrect. So that was one of the big findings of 2025. But we're not done yet. We need to move on. And the next exciting discovery was that of a Czech language publication. And this was called the American Narodny Kalendar or Kalendar. It was published in 1916 by the August Geringer Publishing Company, and that was based in Chicago. Somehow the original language publication disappeared but then reappeared on a digital archive in Czechia called scriptum.cz. So I was so delighted to find this original version of the Czech language publication. Inside that original language publication was a treasure. Over 140 photographs of the Czech victims of the Eastland disaster, most of which have never been seen except for that time when they were published. And I found a place for them, a place where they will most likely be seen. I added these photos to these people's profiles on Find a Grave. There they have an Eastland Memorial site, and I've talked about this many times, and I will put a link in the show notes so you can see it as well. Let's continue. So what I am going to do now is share a commentary by two guests who listened to my episode and they had a lot of insights and they had a lot to say about it. These two actually don't have names, but I call them Ava and Max. They don't exist in real life, by the way. They exist only on Google's Notebook LM, and that is yet another AI tool that a lot of genealogists use and have used. I've used it in the past, and in the last couple of months or so, a lot of people in the genealogy groups that I'm part of were raving about Notebook LM, how much it has improved, how many features it has added. Well, I thought maybe I need to take a second look at this. And I will play the commentary for you and see what you think. And what I hope is that you will be inspired to experiment with it yourself and see what it can do. Throughout the last few years, I've learned a lot from AI, only because I keep taking classes and I've learned to collaborate with it. So without further ado, take a listen to what Max and Ava had to say about episode 147, My Retrospective for 2025, and it's called The Search Goes On. Coincidence, Clarity, and Resolve. Here are Ava and Max.

Eva:

So today we are diving deep into a really overlooked moment in American history. It's the 1915 capsizing of the SS Eastland, right in the Chicago River.

Max:

Right. And our source material for this is, well, it's pretty unique. It's a personal retrospective from a citizen historian, an author named Natalie Zett.

Speaker 2:

And she's detailing these huge breakthroughs that she made just in her 2025 research year.

Speaker 1:

Aaron Powell It's really a mission in uh historical reevaluation. We're looking at how focused, you know, dedicated genealogical work, often done outside of the usual structures, can actually challenge and correct what we've thought were historical facts. In this case. Exactly. We're seeing how one person's effort can completely disrupt the story of a mass tragedy.

Speaker 2:

Aaron Powell And what she found is just extraordinary. We're going to get into how her work corrected the death toll, a number that's been accepted for years.

Speaker 1:

Aaron Powell And the recovery of over a hundred lost photographs of victims, which is just incredible.

Speaker 2:

Aaron Powell But maybe the most fascinating part is her methodology. She attributes a lot of her success to what she calls serendipitous discoveries and even uh the supernatural. So, okay, let's start with the people who actually do this work. Aaron Powell Right.

Speaker 1:

Let's talk about the citizen genealogist or citizen historian.

Speaker 2:

Aaron Ross Powell Yeah, because there's this tendency, right, for institutions to just label them as amateurs and kind of dismiss what they do.

Speaker 1:

Aaron Ross Powell There is. But their value is well, it's just skyrocketing. The thing is, there's this tension between speed and depth. Institutions, you know, they need to cover broad scope, they need peer-reviewed processes, and that takes time and money.

Speaker 2:

Aaron Powell So they can't go deep on everything.

Speaker 1:

Aaron Powell They can't. And our sources point to this. Like there's a piece from The Guardian UK called the Hidden Engine Room, and it notes that these data sets, these uh private passions, are now a gold mine for researchers everywhere.

Speaker 2:

Aaron Ross Powell A gold mine. So these citizen historians are actually building the foundation that the bigger institutions then use.

Speaker 1:

Aaron Powell In many cases, yes. They're creating the data layers that institutions often use, but rarely have the resources to fund or create themselves.

Speaker 2:

Aaron Powell I guess that singular focus is the key advantage then.

Speaker 1:

Aaron Powell It is. It's the core advantage. They don't have shifting priorities. They could afford to spend years on a single question, a single family name. Until it's solved. Right. They're the ones diving into obscure local archives that aren't indexed because they don't have a publication deadline breathing down their neck. They're basically filling in the holes that are left by institutional fatigue.

Speaker 2:

That makes perfect sense. They have the luxury of persistence.

Speaker 1:

Precisely. And in Zet's case, that persistence is paired with a very unusual research methodology.

Speaker 2:

Okay, let's get into that because it feels like a really crucial piece of this puzzle. She says that when her research expanded beyond her own family, the coincidences just started to escalate.

Speaker 1:

Yeah, she called the work, unlike any other research she'd ever done, full of these convergences that she groups under the umbrella of supernatural.

Speaker 2:

Which, you know, for some people that might sound a little out there.

Speaker 1:

It might. But you have to understand her background to get her approach. She grew up with Eastern European immigrant relatives. And in her family, there was this deep respect for, but a healthy fear of the departed.

Speaker 2:

So talking to the dead or acknowledging them wasn't unusual.

Speaker 1:

Aaron Ross Powell Not at all. It was a regular part of life. So this isn't a typical scientific framework, obviously, but it totally shapes her approach to the research itself. It informs this deep listening she talks about.

Speaker 2:

Aaron Powell So for our listeners who might be, you know, purely evidence-based, how do we square that? How do we reconcile these concrete findings, like a new death toll, with a method that relies on serendipity?

Speaker 1:

Aaron Ross Powell That's a great question. And I think it comes down to the hidden tools of investigation. When she talks about deep listening, what she's really describing is a process of infinite patience, of being open to clues that don't seem linear. Trevor Burrus, Jr.

Speaker 2:

And pattern recognition.

Speaker 1:

Intense pattern recognition. And this is where her other skills come into play. She's a published writer, an investigator, and an IT consultant.

Speaker 2:

Aaron Ross Powell, right, we were going to get to that. The background in designing and testing systems.

Speaker 1:

Aaron Powell Exactly. So think about it. When a weird clue appears in an unexpected archive, an academic might just dismiss it as a random event, an anomaly.

Speaker 2:

Aaron Powell But she sees it differently.

Speaker 1:

She sees it like a system fault. If a system is suddenly giving you a strange result, you don't ignore it. You follow it. Her heritage just gives her the framework, the motivation to never stop following that lead, no matter how irrational it might seem.

Speaker 2:

Aaron Powell And that's what led her to break open the fatality count.

Speaker 1:

Aaron Powell That's what did it okay?

Speaker 2:

So let's move to that finding. This is the biggest historical correction she made. She called it the million-dollar question. How many people actually died on the Eastland?

Speaker 1:

Aaron Powell And the central, frankly, shocking revelation is that there is no source-sided, peer-reviewed, official list of the fatalities.

Speaker 2:

Aaron Powell Wait, for a disaster this big, there's no official list.

Speaker 1:

The number everyone uses 804 to 4 was always, always an estimate.

Speaker 2:

Aaron Powell And where did that number come from?

Speaker 1:

Aaron Powell It came from George Hilton's book, Eastland Legacy of the Titanic. But here's the crucial part. Hilton himself, in his own appendix, called it an estimate. And he actively encouraged other people to continue the research.

Speaker 2:

Aaron Powell So what happened? Why did everyone stop?

Speaker 1:

Aaron Ross Powell Well, the original Eastland Memorial Society, they disbanded in the early 2000s. When they did, all the momentum for verification just it died. Hilton's number got repeated so many times it wasn't a statistic anymore.

Speaker 2:

It became a brand.

Speaker 1:

So it became a brand, yeah. It was convenient. And that, ironically, is what stopped the research cold.

Speaker 2:

Aaron Powell So Zette comes in in 2025 and challenges this brand directly. And she finds two major corrections. The first one was a man named Thomas Marin.

Speaker 1:

Yes. In October 2025, she confirmed his loss. He was just inadvertently omitted from Milton's original research. The records were just incomplete.

Speaker 2:

Aaron Ross Powell And with his addition, the estimated loss of life immediately jumps to 845.

Speaker 1:

Right. And she published that finding immediately, which really shows you the urgency these citizen historians feel about correcting the public record.

Speaker 2:

Aaron Powell And then came the second correction, which is, I think, a lot more complicated. Chrissy McNailitz.

Speaker 1:

Yes. This one is much more ethically complex. She didn't die on the day of the disaster. She died seven years later. But her family always said she got sick right after being rescued, and in their words, never got better.

Speaker 2:

Aaron Powell So this just cracks open a huge question about what we even define as a casualty in a mass tragedy.

Speaker 1:

It really does. Adding her brings the adjusted estimate to 846, but it forces a conversation about, you know, proximate cause of death.

Speaker 2:

And the standards for that?

Speaker 1:

Exactly. I mean, should a disaster's death toll only count the people who died that day? Or should it include someone whose death years later can be proven to be a direct result of injuries or trauma from the event?

Speaker 2:

Aaron Ross Powell And right now there are no official standards for the Eastland.

Speaker 1:

None. No scholarly consortium has ever set an evidence-based standard. So until they define the terms, any number, even 846, has to be called an adjusted estimate. It's an administrative failure, not a research one.

Speaker 2:

So moving from the numbers to the individual stories, this is where her work gets really intense. She starts tracing the death of a woman named Julia Kolar.

Speaker 1:

And this is a perfect example of her skills coming together. She finds Kolar's death mentioned in this obscure trade journal, and that leads her to a translated article called Catastrophe on the Chicago River.

Speaker 2:

And that article detailed every single Czech person who died.

Speaker 1:

Which confirmed that the best, most comprehensive reporting at the time had been done in a non-English source.

Speaker 2:

Trevor Burrus And this is where her tenacity kicks in. She wasn't satisfied with the translation. She said she was hell-bent on finding the original 1916 Czech publication.

Speaker 1:

Aaron Powell, which is a massive needle in a haystack. I mean think about the hurdles. You've got language barriers, different search indexes. A lot of these old journals are just scanned PDFs you can't even search.

Speaker 2:

But her IT background let her get past all of that.

Speaker 1:

It did. She found the journal on a digital archive in Czechia.

Speaker 2:

And what she found inside, it's just staggering.

Speaker 1:

It is. The original 1916 publication contained over 140 individual original photographs of the Czech victims.

Speaker 2:

Wow. Over 140 photos.

Speaker 1:

Most of which had not been seen, had not been published since 1916. They were literally one bad hard drive away from being lost forever.

Speaker 2:

That's phenomenal. But finding them is one thing. How did she get them out there so people could see them?

Speaker 1:

This is what she calls the Find a Grave marathon.

Speaker 2:

Okay.

Speaker 1:

She knew that Find a Grave is this massive online cemetery resource, and crucially, it's constantly indexed by Google.

Speaker 2:

Ah, so it's findable.

Speaker 1:

Instantly findable. So over three days, she extracted all those photos and uploaded them to the Eastland Disaster Victims Virtual Cemetery on the site.

Speaker 2:

That's brilliant. I mean, it completely bypasses the years it would take for a museum to digitize and publish that kind of thing. She used a public tool for immediate impact.

Speaker 1:

Aaron Powell And you can't overstate the human impact of that. She told us one story about adding a photo to a victim's profile, and a great granddaughter left a message saying she had never seen her photo.

Natalie Zett:

Wow.

Speaker 1:

It transforms a name on a list back into a person. And then that's not just emotional, it's intellectual. It provides real data for families, for academics, for generations to come.

Speaker 2:

Aaron Powell It's clear that this level of work required, you know, a pretty serious skill set to begin with.

Speaker 1:

Oh, absolutely. Her professional background is the engine room here. Publish writer, investigator, IT consultant. She knows systems. She applied project management and forensic data analysis to a history problem.

Speaker 2:

Aaron Powell And she also recently took public history classes at the University of London, right? So she's merging that citizen approach with a more formal academic one.

Speaker 1:

Aaron Ross Powell She is. Oh. And she grounds it all in this deep respect for her subjects, a model she says she learned from Lisa Louise Cook of Genealogy Gems.

Speaker 2:

Aaron Powell What did she say about that?

Speaker 1:

Aaron Ross Powell She credited Cook for giving her personal hands-on care and for treating her family's history as if they were her family. And that mindset, you know, that these subjects deserve full respect and investigation, that's why she won't settle for an unverified number.

Speaker 2:

Aaron Ross Powell You know, this whole deep dive into her 2025 research, it really confirms a crucial lesson about how information becomes history.

Speaker 1:

Aaron Powell It really does. It confirms that history is never really done. It's more about who keeps looking. The Eastland story was closed prematurely because of, well, institutional fatigue and the convenience of a number that was easy to repeat.

Speaker 2:

And her work just shows that one dedicated person using modern digital skills can successfully challenge and correct a century-old statistic.

Speaker 1:

Accuracy can always win. It just takes persistence.

Speaker 2:

And that leads to the final provocative thought that Zet offered for 2026. She's predicting a kind of moment of reckoning.

Speaker 1:

A challenge, really.

Speaker 2:

Yeah, a challenge to the institutions that have used the old number 840 for as a brand. They now have to face this new confirmed evidence. So the question is, will they change their official counts?

Speaker 1:

Or will this correction just slowly disappear? Because changing it is administratively inconvenient.

Speaker 2:

It really puts the historical gatekeepers on the spot, doesn't it?

Speaker 1:

It's a moment of accountability for them for sure.

Speaker 2:

Aaron Powell And we hope that for you listening, you use the story as encouragement to take a second look. At everything you've heard about this or any other historical event, if you see a number or a fact that just feels a little too tidy, ask yourself, who established this? And when was the last time anyone actually verified it?

Speaker 1:

Because in the end, the goal is always to make sure the historical record matches what really happened.

Speaker 2:

Even when the truth is much, much harder to find.

Natalie Zett:

It's like having someone hold up a mirror, except that someone holding up the mirror is made of code. It shows you things that you didn't necessarily realize at the time when you're writing something. This tool has become more sophisticated. And did they get everything right? Not necessarily, but they got more right than anything else. The other thing that they were able to deal with was the nuance. The history of the people of the Eastland disaster, like all history, like all genealogies, it's nuanced. It's not a neat narrative that should be controlled by either the repetition of a number or by the repetition of the same stories over and over again. There is so much more to the stories of all of these people who were involved with the Eastland disaster. And it's only been in the last couple of years, as I've done this deep research, have I realized how rich, how engaging, and how fascinating it is. And also I've realized that I've only scratched the surface. There's a lot more to be discovered. And it's true that all of this is messy because, as Krista Collins says, families are messy. And there's a beauty in that because we are all human. And telling the real stories of each one of these people helps bridge that gap. We can see ourselves in the people of the Eastland disaster. And if you have something like this in your background, there are reasons to rejoice because these people, this history can teach you so much about what's happening now, not only what happened then. And in a way, certain themes continue to repeat throughout history. And knowing sometimes that our ancestors went through similar things as we are going through now, that can be strengthening and can be helpful. And it can give us both wisdom and perspective. So I won't talk longer about this, but next week I'll be back and we will continue reading from the article Catastrophe on the Chicago River by Joseph Mach. And it is a riveting, riveting story. In the meantime, remember to take care of yourselves and take care of each other. And we'll talk next week. Hey, that's it for this episode, and thanks for coming along for the ride. Please subscribe or follow so you can keep up with all the episodes. And for more information, please go to my website. That's www.flowerinthher.com. I hope you'll consider buying my book available as audiobook, ebook, paperback, and hardcover because I still owe people money, and that's my running joke. But the one thing I'm serious about is that this podcast and my book are dedicated to the memory of all who experienced the Eastland disaster of 1915. Goodbye for now.

Head Shot

Natalie Zett

Host