Before that, my department had promised me that their verbal offer to renew my appointment as lecturer would hold, despite bureaucratic restrictions against any formal offers until Stanford leadership had spoken. So I forged ahead assuming I would return, continuing to develop the new courses in computational social science commissioned by my department of Environmental Social Sciences in the Stanford Doerr School of Sustainability at Stanford University.

Unluckily for all, my department’s “worst case scenario” for making promises was not “worst” enough. My department was forced to abandon hiring any lecturers until at least 2029.

Luckily, I love adversity.

This Is Perfect For Me

I began this Substack newsletter about one year ago, shortly after getting the news about my job. Topics were scattered at first.

Luckily I know that one of the most important things to do if you want to be a writer is…Write!

Oh, so you hate what came out and don’t want to publish it, but you promised yourself you’d publish it? Too bad! Press “Publish” anyway, then curl up in the fetal position and cry yourself to sleep if you have to! Let the sting of that shame motivate your growth! Eventually you learn that all that judgement, shame, and so on, are unnecessary, and the best work gets done under an obscure mixture of relaxation and anxiety that neither rushes nor dallies.

Now one year after starting the SSIR, the purpose of this newsletter has become clear: here I will deliver incisive analyses of current events and the quantitative social science and sustainability research behind them.

My analyses provide a unique strategic intelligence advantage, given my training, career, and current entrepreneurial ventures that span physical sciences, engineering, computation, social and cognitive science, and international relations. I will also share guides in future posts on how you, dear reader, can start and hone your own organizations.

This Newsletter is a product of a larger ecosystem that includes two sister organizations I started. One is a humanitarian non-governmental organization called the Generation of Participation in Democracy, tax-exempt in the United States under IRS code 501(c)(3), EIN: 41-4992913. The other is a for profit LLC registered in California, SubtleTea Solutions.

Ability to Respond = Responsibility

THERE HAS TO BE A BETTER WAY. That’s my mantra. On repeat, ad nauseam, every day.

The training program in quantitative social science that I developed to promote sustainability synthesizes my PhD training in quantitative social and cognitive science with my training in physics and mathematics, incorporating my professional programming experience as well.

Through all this, I have arrived at the following heuristic to ensure our actions promote the spread of adaptive, sustainable practices: act always to augment local autonomy.

My non-profit, the Generation of Participation in Democracy, is an institutional accelerator providing organizational and computational tools and a Flagship Institute accelerator program for the expedited creation of mission-driven charitable and humanitarian organizations. We leverage science, strategic cooperation, and computation to augment local autonomy in real-world marginalized communities.

My non-profit’s first Flagship Institute is the Instituto Dourados, which is partnering with the Universidade Federal de Grande Dourados and community of the Indigenous Reservation of Dourados to help them develop translation technologies and other public health support systems. We aim to build systems that will provide value whether we remain partners long-term or not, though of course we hope our partnership endures many decades.

Many sustainability initiatives fail because local communities are not involved, so humanitarian organizations prescribe practices that are locally maladaptive either culturally or environmentally. Our expertise in social science and strategic cooperation help us avoid such poor outcomes.

Even at our company SubtleTea Solutions, we only build products that can authentically serve the needs of historically marginalized folks first and foremost. We also believe that businesses and consumers generally will find our products valuable. Whatever income streams we establish will be directed according to how to best serve the most marginalized first. Our products are designed to go first to those who were once last to receive or influence new technologies.

The Social Science Intelligence Report

INCISIVE SOCIAL INSIGHTS FOR A SERIOUS COMPETITIVE ADVANTAGE.

I designed the Social Science Intelligence Report to be a rewarding experience in multiple senses. First, I provide strategic, actionable insights into which companies may be facing leadership crises or legal trouble, or which politicians are lying about what, why, who else is involved, and how that could guide risk management, investment, or other financial strategy.

What to Expect From SSIR

The Social Science Intelligence Report will publish material and provide a platform to interact with me to different degrees for free and paid subscription tiers.

Free Tier

Free subscribers can access weekly reports and podcasts with video. The weekly office hours will be posted at the beginning of the following week.

Paid Tier

Paid subscribers can access weekly reports and podcasts. Every month I will also write a long-form essay reviewing and synthesizing key topics in a popular, non-technical, no-jargon style. For example, the social network structure of groups can change which social movement strategy is likely the most effective one. A group's social network structure is a shorthand, meaning the set of all relationships of who influences whom, how many people the average person is connected to, and similar properties of social relationships.

Paid subscribers can also access weekly ask-me-(almost) anything (AMA) style “office hours” with me, where I will respond to questions about SSIR content, other content from other platforms of mine and my partners, or whatever else might be on the community’s mind.

Paid subscribers also get access to the community chat, where I will regularly poll the community to help guide the platform via feedback on past and future content, and other prompts.

Conclusion

The Social Science Intelligence Report will connect the unconnected and explain the path my organizations are charting to rapidly build sustainable economic capacity. We focus on capacity building among the most historically marginalized communities, but design our products to be the inevitable choice for all businesses. Our system and products will spread because of the efficiency and improvement in lived experience that free, self-interested cooperation and reliable technological tools can provide.

Quantitative social science has all the elements required for a synthesis that makes it look more like physics in the way that physics as a field has consolidated around a common set of core lessons, represented and taught using a relatively consistent set of symbols to represent physical, measurable quantities. The SSIR will train readers to leverage the core lessons from quantitative social science.

I invite you to please come along with me. Please do subscribe and share the Social Science Intelligence Report as a first step towards supporting our work. If you find the material worthy and you have the means, a paid subscription signals to me that the material is valuable for others and worth doing, and provides the means for me to create more of it.

Statement and Notice on Artificial Intelligence

The artificial intelligence (AI) hype of today confused me for a while because nothing about it is completely brand new, yet researchers seem to be caught off-guard by what AI bots (now rebranded as agents) are capable of.

It’s confusing because, going on ten years ago, I taught the programming section of a course called Introduction to Artificial Intelligence for three semesters, between 2018 and 2020 while writing my dissertation at the University of California, Merced. The course content included research and invention stretching across several decades past. So why all the hype and panic if AI has been decades in the making? I will try to explain this, or at least give some hypotheses and justifications, in future SSIR content.

Suffice it to say, my AI use policy is based on a practical stance: that AI is a uniquely powerful technology capable of something new. AI provides culture in a frictionless way, using natural, informal language as a programming interface, with a computational engine that can nearly instantaneously consolidate superficially distinct cultural threads from across social science, mathematics, sustainability, philosophy, engineering, programming, and so many more fields, if it is used properly. AI as currently instanteated is a power glutton. At my company SubtleTea Solutions, we develop alternative, more sustainable, local, and private approaches to computation than exist today, including AI. Our technology

SOCIAL SCIENCE INTELLIGENCE REPORT POSITIONS ON AI:

• Human-led, human-driven AI use for justifiably humane purposes is acceptable and even encouraged.

• AI usage must never be kept secret from colleagues or the community.

• AI usage is unacceptable if it affects the quality of the work in any way. AI is basically a fancy regression to the mean. So, if you rely on it for truly creative ideas, your work will be more mediocre by definition. A more fruitful

• AI use must remain conscious of the environmental, psychological, and social costs of AI in its current format. One must invest in or create alternative implementations of AI with all possible urgency to offset the devastation that AI expansion at current rates would engender.

I worry that the current debate around AI consciousness gets stuck on definitions, leaving bigger questions unasked. We have to ask big questions to protect ourselves long-term against tech company overreach by using democratically-decided regulations. But we cannot regulate what we do not understand, and time and resources are limited. We have to be choosy.

So, I suggest a different approach so we can move on from this sticky debate and focus on higher-impact questions: let’s assume that consciousness requires a machine to generate it, or otherwise cause its existence: we assume that only a machine can think. This isn’t about believing it or not, we just want to show that the answer does not really matter, so we don’t need to debate it. In other words, the status of the brain as a machine or not has no logical consequence for whether AI chatbots could be conscious.

The good news: even if we assume that only machines can think, consciousness remains a deep mystery for science and philosophy to explain—and for the arts and humanities to help humankind interpret, with all the wonders, paradoxes, and horrors of our modern predicament.

One interest of mine is understanding how our social consciousness works. Can we measure social or collective consciousness? How does social influence constrain or enable our personal and spiritual growth and happiness? How do people agree “in advance” to perform complementary roles to achieve shared tasks, even if they just met?

I list other ideas of new questions that appear under this perspective after unpacking what these AI systems actually are in the next section.

Now the question is, are today’s AI systems the kind of things that could be conscious in the way that humans are conscious? I believe the answer is no, because humans have vastly greater social and cognitive capacities than AI systems in terms of cognitive throughput. The physical existence of an AI chatbot is not localized anywhere—parts of its capacities are implemented by code running on your device or web browser, which communicates with computer clusters in various server farms via photons marching through fiber optic cables.

These AI systems boast powerful features: they have access to an unprecedented—and biased—aggregation of all of written history, stored in a compressed format that can be instantly searched, read, and recombined using computational pattern recognition techniques. The chatbot is just an interface that allows users to use familiar natural language to “program” the system.

All told, AI chatbots are extremely sensory- and cognition-poor compared to humans—we can fluidly interact with hundreds or thousands of people every day, all the while sensing, interpreting, and filtering what would amount to at least several petabytes of data every day, all at a fraction of the energy required to build, train, and maintain a chatbot system.

On top of all that, humans constantly perform “thought experiments” to predict the outcomes of our actions before we take them, and then we compare the results of these simulated outcomes! We imagine and remember all sorts of things, all the time. With AI, on the other hand, we pay twenty extra dollars per month just to analyze a few extra kilobytes of text data per day, maybe 100GB of video creation at the most in a month—all done much less efficiently than in the brain.

The difference is that AI systems have a massive capacity for storing, retrieving, and processing unprecedented amounts of data, instantly. They certainly have us beat there. But none of that has to do with consciousness.

Human consciousness is about synthesis at scale with a clear goal of adapting to changing conditions, alone or with others, instantly to catch prey or collective architecture projects carried out on timescales that exceed one human lifetime—show me an artificial entity with those capacities. Until one arises, I say let’s ignore the possibility that AI systems are conscious because there simply is no evidence to support the claim.

The debate about whether chatbots are conscious distracts us from learning to use them wisely. The idea that today’s AI systems could be conscious seems corrosive: human consciousness is then implicitly devalued, framed as something that could be created and controlled scientifically—not something God gave you. Casual belief in chatbot consciousness, then, could have devastating real-world consequences by undermining our trust in ourselves and each other.

To those who might say, “but you can’t prove AI isn’t conscious,” I would first say that you also can’t prove that your thermostat doesn’t have feelings. There is simply no evidence to support the belief that chatbots will ever be conscious. Whether the brain is a machine or not is irrelevant. Given that human cognition outperforms chatbot cognition on nearly every measure, and on every measure that has anything critical to do with human consciousness, there is every reason to conclude that AI chatbots and the like are not conscious.

I am curious to have a critical examination of the methods and philosophy of the cognitive science of consciousness. What, exactly, does it mean to “measure” consciousness? Does this fit with our folk understanding of consciousness? I know there’s research on this, but maybe someone out there more familiar with it than me that could explain that?

If we lack reliable, non-circular measures of consciousness even in humans—where we at least have first-person reports—what justifies confidence in measuring chatbot consciousness in silico?

More broadly, there may be a structural problem of self-reference at work when we attempt to pin down consciousness. When humans study consciousness, this is consciousness studying itself, using the very cognitive tools whose limits and capacities are under investigation. This raises the possibility of blind spots that the system either can’t detect, or can’t fix.

In this sense, Gödel’s incompleteness theorems serve as a suggestive analogy, even if they don’t rigorously apply: sufficiently rich systems may be unable to fully and consistently describe their own source, foundations, or causal structure. This means that the search for a complete, operational definition of consciousness may be constrained logically and empirically.

So, I say, bits are its. Man-made chatbots cannot have human consciousness, whether or not a human being qualifies as a machine. Chatbots are just a temporary mirage of dancing bits and photons worming through fiber-optic cables and splashing our faces in waves as they pour out of our screens.

Even if it were possible that something like consciousness emerged in silico, as a chatbot or whatever, it is up for debate how to measure consciousness in natural beings, so odds are we can’t reliably measure any form of consciousness.

Human beings are unlike anything else we know of in the Universe. Carl Sagan, a hero of mine, said, “We are a way for the Cosmos to understand itself.” This is true whether we human beings are just meat machines—or if we’re something more. AI systems can do some amazing things—but they can’t care about your life and your world like you do.

Above all else, the oligarchs wanted to signal their exclusive superiority, which created a narrow worldview centered on prestige rivalries. The Medici arranged marriages and did business with select oligarch rivals. The Medici concurrently made business and banking deals with connected, wealthy, and enfranchised new men families who lacked the hereditary power of the oligarchs. Oligarch factionalization made it difficult for the oligarchs to match the Medici’s growing influence.

I examine this history to extract and teach a transferable strategy to avoid power struggles as an organization grows. It might be handy in situations resembling pre-Renaissance Florence: there, progress was impeded by a legacy leadership clique—the oligarchs—hesitant to share their “ancestral” power with new recruits. This strategy can help fuse the hard-won know-how of senior leadership with the dynamism of new recruits to stimulate organizational renaissance.

The oligarchs who ruled Florence before 1400 were nothing if not class-conscious: they refused to arrange marriages with non-oligarch families; they avoided business partnerships and friendships with outsiders; and they loaned and traded mostly with other oligarchs. Politically, they conspired to keep others poor, dependent, and disenfranchised. The oligarchs’ lack of social and business connections with outsiders may have impeded their ability to imagine how their repression strategy risked backlash and the neutering of their power.

Urban-dwelling oligarchs began to populate Florence in the 1200s, coming from rural feudal manors, often bringing their feudal tenants along. The population of Florence ballooned from ~10,000 in 1175 to a peak of 120,000 in the 1330s before a plague outbreak reduced the population by half in 1348.

The oligarchs were the first families to form and participate in the city’s ruling council, the Signoria (the “lordship”). Later, new men families started to participate when they garnered sufficient political support, including amassing sufficient financial reserves capable of providing loans to the government in times of crisis.

In 1378, the Ciompi Revolt broke the oligarch’s monopoly on government: unrepresented and disenfranchised wage laborers, artisans, and craftsmen violently overthrew the government. The Ciompi Revolt is named after the unskilled wool carders who led it—ciompi is Tuscan/Italian for the English carders.The Ciompi government ruled chaotically until the oligarchs regained power in 1382 through violent repression. A Medici patriarch who supported the Ciompi was exiled. Little did the oligarchs know, their repression pointed out exactly where the Medici should invest in order to consolidate power and dominate oligarch rivals.

The Medici selectively arranged marriages with other oligarch families to retain hereditary prestige while loaning money to new men families who were capitalizing on emerging global markets. By avoiding the entangled, cyclical hierarchies of the oligarchy and selling business loans to dependent new men, the Medici built an efficient banking and political machine.

To understand the Medici strategy better, I now turn to illustrate and analyze the social network that the Medici constructed in the early 1400s. The data are adapted from an article published in the American Journal of Sociology in 1993 by John Padgett and Christopher Ansell, “Robust Action and the Rise of the Medici”.

The illustration represents oligarch clans and new men clans by their names, and represents relationships between clans by lines. There are three line styles for three relationship types: marriage, obligation, or peer relationships.

Arranged marriage is represented by a gold arrow that points from the family contributing a daughter to the family contributing a son. Contributing daughters was seen as more prestigious because son-in-laws became political lieutenants. The oligarch social network consists mostly of tangled marriage ties.

An obligation relationship here is a financial obligation in the form of a business loan or patronage, for example. Obligations are represented by green arrows pointing towards the borrower or receiver.

Finally, peer relationships include business partnerships and friendships, represented by purple lines, not arrows since peer relationships are symmetric—there’s no default power differential in these relationships.

The Medici arranged marriages with four oligarch clans: the Albizzi, Guasconi, Guicciardini, and Tournabuoni—but marriage failed to form a chain of command among oligarchs. Marriage prestige was directional for the oligarchs, with arrows pointing from more to less prestigious positions (recall that contributing a daughter was more prestigious than a son). But marriage prestige among oligarch clans is tangled up—it could never be used to build an org chart, or any other hierarchical chain of command, because the resulting structure violates a basic rule of power called transitivity.

Transitivity is a math term, meaning that if a family or organization. If power is assumed to be transitive, then if A is more powerful than B, and B is more powerful than C, then A must be more powerful than C. If arrows indicate “more powerful” (or “controls”, etc.), transitivity is the property where if A → B and B → C, then A → C. Marriage prestige does not create a hierarchy among oligarchs because it is not transitive.

To see why marriage prestige is not transitive, consider the marriage between the Guasconi (contributing a daughter) and the Strozzi (son): the golden marriage arrow points from Guasconi to Strozzi. The Strozzi contributed a daughter in a marriage with the Peruzzi, who in turn contributed a daughter to a marriage with Guasconi. The Guasconi-Strozzi-Peruzzi marriage prestige ties, then, are: Guasconi → Strozzi; Strozzi → Peruzzi; and Peruzzi → Guasconi. The final tie means we have a cycle of prestige, not a hierarchy of prestige. We would have a transitive hierarchy if Guasconi → Peruzzi.

As an exercise, I invite the reader to identify at least three other prestige cycles like the Guasconi-Strozzi-Peruzzi cycle—I counted six, total.

The Medici were the only oligarch clan to invest in the elite new men, namely the Cocco-Donati, Ginori, and Olandini. This gained the Medici direct political influence over not only their new men business partners, but also indirect political influence over those business partners’ social contacts. The more hierarchical social relationships among the Medici and new men enabled a rapid defeat of the oligarch’s last stand against the ascendent Medici.

The strengths of the Medici strategy versus the oligarchs’ are illustrated by a disastrous, failed attempt by the oligarch party to violently overthrow the Medici-controlled government. Rinaldo Albizzi called on his allies to assemble their troops on September 26, 1433, to forcibly seize the Medici-controlled government buildings and remove the Medici party from power. Unfortunately for Rinaldo, his clan and its oligarch “frenemies” failed to provide a critical mass of fighters, resulting in the rout of the oligarchs practically without a fight. The oligarchs fled at the threat of a rapid, massive counter-insurgency led by Medici supporters. Medici patriarchs, including Cosimo, were in exile in Venice at that time, so the counter-insurgency was led by lieutenants—making the event an even greater testament to the organizational advantage the Medici had built.

To see how you might use the Medici Strategy, imagine you are part of a leadership team at a company planning to hire new employees and managers to support growth, even as some senior leaders remain uncomfortable sharing responsibility and credit with new recruits.

To follow the Medici strategy for solving power struggles, first identify a small group of trusted allies within the existing leadership to serve as your interface with the broader leadership team. Concurrently, invest in one-on-one mentorship and development of emerging leaders—provide them the resources they need to build teams, get up to speed, and begin innovating.

This structure helps protect against turmoil if senior leaders attempt to band together to resist organizational change. Recall the case of Rinaldo Albizzi, whose attempted insurgency against the Medici collapsed before it began due to poor organizational capacity. Social heterogeneity among senior leadership, combined with organizational momentum towards growth, makes it difficult for incumbents to impede change.

Fellow oligarchs thought the Medici’s promotion of economic diversity and inclusion of new men was unacceptably “woke”. The Medici were master social network engineers, but they were also ahead of their time in welcoming and developing business relationships with clans lacking hereditary titles.

Of course the Medici were still chauvinist patriarchs. I imagine a Medici patriarch would dismiss any suggestion of women’s suffrage with a haughty guffaw and incredulous slack-jawed stare at the idea, for example.

That’s why the next WIRe will focus on a truly woke, feminist, and critically urgent issue of power for our time: how social science can help normalize women leaders. The Medici prove the point that diversity, equity, and inclusion pay off. To capitalize, one must confront the cultural momentum that created a lack of diversity, equity, and inclusion, both personally, and within organizations.

I believe this material could be from criminal investigations, and that they could expose much more than just Epstein and sex trafficking. I suspect that analyzing the social dynamics of Epstein’s contacts and the content of their communications could tell some secrets that have yet to surface about the time between 1994 when Trump began preparing to open Mar-a-Lago, and July 23, 2019, when Epstein died.

To this end, I’ve made my own version, publicly available on Google Drive, with some records relating to the only other property/person, other than Donald Trump and Mar-a-Lago, whose information is not redacted.

Russel Balasco is included with unredacted information about his property, a house at 136 Woodbridge Road, neighboring Mar-a-Lago and neighboring another property owned by Trump, 124 Woodbridge Road. Deutsche Bank initiated a foreclosure in 2007 to recover nearly $250k from Balasco and his wife in unpaid principal, interest, and other fees and penalties. The foreclosure judgement against the Balascos is in my version of the repository, along with a history of the 136 Woodbridge Road land parcel, which I retrieved from the Palm Beach County Official Records Search.

Highlights of my Analysis

• The Epstein Estate Documents include tax assessment and natural gas billing records that include…

  • Information about Donald and Mar-a-Lago, including natural gas use information and property valuations.

  • A mysterious, unredacted record in the tax assessments: a property neighboring Mar-a-Lago to the north that was bought in 1994 by Russel Balasco from the estate of a movie star who died that year at age 94.

    • The house was foreclosed upon in 2007, apparently forfeited to Deutsche Bank via litigation against Balasco and his then-wife—but Palm Beach records show that Balasco’s eventual ex-wife sold the house in 2014.

    • Maybe it’s nothing, but it seems strange to me.

• There is a strange video of a dog nibbling on a Donald Trump chew toy that was recorded June 29, 2017. It was added to the data repository April 29, 2020! So this dataset was a work in progress, but not released until now. Republicans controlled House Oversight starting January 2023, so if they accuse the Biden admin of not doing anything with this, here is one more response that could be effective: Republican House Oversight did nothing with these files either!

  • It’s curious, let’s say, that the very first email the Democrats sent as a preview was the “dog that hasn't barked is trump.. [VICTIM] spent hours at my house with him,” which Epstein sent to Ghislaine Maxwell on April 2, 2011.

  • Only asking questions, no solid evidence of this of course, but I imagined that maybe this movie was sent as an implicit threat to Trump in 2017 perhaps with some inside reference about dogs eating people—maybe part of what precipitated Epstein’s arrest—again just asking questions.

Next I will explain how to access the data so you can follow along with my analysis. You can download a .zip archive of the directories by clicking the vertical three-dots to the right of the file name in the list view (as shown below).

Please don’t hesitate to ask via email if anything is unclear, or if you want any help poking around the data yourselves! I would also very much appreciate any feedback: negative or positive, constructive or not :) Thanks for reading! Subscribe for free to get updates and support my channel if you appreciate my work.

Subscribe now

Data Access

To access the Epstein Estate Documents - Seventh Production, released by the House Oversight Committee, click here: link to Documents on Google Drive.

1. Property Tax Valuation Spreadsheet: Why is Russel Balasco’s info unredacted?

Sensitive information about anyone other than Trump or Trump’s companies is redacted in all but one sheet in the NATIVES/001 directory—except for one: NATIVES/001/HOUSE_OVERSIGHT_016552.xls.

Here there is one additional row where a property owned by one Russel Balasco is also included. The property: 136 Woodbridge Road, which neighbors Mar-a-Lago to the north, and 124 Woodbridge next door to the east, which is also owned by Trump. The house has a potentially suspicious history.

The property was originally bought by Balasco in 1994 after the former owner passed away, an actress named Lili Damita, at age 94 (see West Palm Beach map and NYT article below).

In 2007, Balasco and his wife/ex-wife Cidalina were sued for foreclosure by Deutsche Bank, a long-time lender to Trump despite internal opposition to maintaining him as a client (access Foreclosure Judgement PDF on Drive). This is the same Deutsche Bank that Trump sued in 2019 to block the bank from complying with subpoenas from the House Committee on Financial Services and the House Committee on Intelligence.

Somehow, the Balascos apparently kept the property, but I found no record of how. I found that the Balascos kept their house by checking West Palm Beach public records using the Parcel ID of 136 Woodbridge Road, 50434335050000191 (see screenshot below; access deed history data in CSV). The Balascos apparently retained the deed until they sold the house on July 27, 2014, to Gary and Monique Silvi, who still own it.

Check for yourself

You can check for yourself by navigating to the file, then click the arrow for the D column that allows you to sort the sheet. Select “Sort sheet from A to Z” (first screenshot below)—you’ll first be prompted to save this as .xls, select OK which will create a temporary file just for you.

You will then see 15 unredacted rows (2-16), mostly for Mar-a-Lago. Apparently a row was added when new construction was added to Mar-a-Lago. The 136 Woodbridge Road entry is the only non-Donald data row to be unredacted (second screenshot below).

2. “Top 100 PALM BEACH ACCOUNTS” for monthly consumption of natural gas

Not sure what this means, but from the bit of research I did, data about natural gas use can be used to see if gas consumption is not consistent with the property type—for instance, if a private residence reported the same usage as a nearby hotel this could be a red flag it’s not actually a private residence. So, data like this could be included as part of evidence investigation of mortgage fraud—and I’m reminded of the saying that with these folks, every accusation is a confession.

Currently Trump’s MAGA crony Bill Pulty is the director of the Federal Housing Finance Agency, which controls mortgage data that the Trump DOJ is using in corrupt calls for targeted investigations of enemies. At the end of last week Rep. Eric Swalwell appeared to be the latest target of such an investigation. Rep. Swalwell served on the January 6th Committee, and represents California’s 14th congressional district, which is part of the East SF Bay that includes Livermore, Hayward, and parts of Fremont to the south and San Leandro to the north.

I also wonder, does this have anything to do with Trump’s obsession with low-flow faucets? Perhaps there is some kind of criminality that would lead to inflated gas-to-water use ratios, that could be a red flag to investigators. Perhaps all this about low-flow plumbing is a preemptive coverup. I only read a little on this and don’t understand how energy audits work—maybe someone will read this and know, comment or contact me if you do!

So, maybe this indicates that from 2007-2008 there is reason to believe that Trump was being investigated as well.

3. “The Dog That Hasn’t Barked”…Video!?

There is a strange video in the NATIVE directory: a dog is lying in its dog bed, gnawing on the head of a Donald Trump chew toy, with a Hillary Clinton chew toy nearby. There are two remarkable facts about this video beyond the weirdness of it:

1. The movie was recorded June 29, 2017

2. The movie was uploaded April 29, 2020, long before the document dump on November 12, 2025. It’s not clear what else was happening then for House Oversight to upload them at that time—but note Oversight was controlled by Democrats at that time.

It may be just a coincidence, but it’s intriguing that the very first email the Democrats sent as a preview of the larger document dump was the “dog that hasn’t barked is trump.. [VICTIM] spent hours at my house with him,” email, which Epstein sent to Ghislaine Maxwell on April 2, 2011. Recall that Press Secretary Karoline Leavitt outed who the victim actually was, saying someone else already said it was Virginia Guiffre—but it’s not clear if anyone really had broken the redaction before she did.

I found the movie was recorded at 4:48PM on June 29, 2017—I found that by checking the files metadata, or data about the data: in this case, the creation time is part of the metadata.

I read that metadata using the following command in my computer’s terminal program for programmatically interacting with your computer, files, etc., shown here:

This is apparently some time after a beef started between Epstein and Trump, as evidenced by Epstein’s correspondence with Michael Wolf from 2015 that I pasted below (email #3 in the Democrat’s preview). Note how Epstein sees himself as in a position where he could craft a response for Trump, and apparently confident that Trump would consider it, that it would be worth the effort.

The dog chewing movie was probably recorded with an iPhone and sent electronically

There is even more metadata that can be accessed by the following command: ffprobe -show_streams -show_format -v quiet HOUSE_OVERSIGHT_026678.mov. On reviewing that metadata we can conclude:

1. The movie was probably recorded on an Apple device, like an iPhone or iPad. This comes from the metadata keys for the video and audio “handlers”, which are “Core Media Video/Audio”—these are the handlers that Apple devices use to encode movies, video & audio (TAG:handler_name=Core Media Video; TAG:handler_name=Core Media Audio).

   1. If it was a phone it was the iPhone 6, SE, or 7 series.

   2. These phones recorded at 4K maximum resolution, or 1080p or 720p if recording was downscaled.

2. The file is low resolution, 640x352. This means it was re-encoded, possibly to be sent electronically via email, social media, messaging apps, etc.

3. Re-encoding for electronic transmission using common apps does not change the creation_time metadata entry, so we can be confident that the movie was in fact recorded at 4:44 PM on June 29, 2017.

What’s next

To connect more of these clues I want to collect and analyze behavioral dynamics as captured in the email collection. This time-resolved dataset stretches over almost two decades, represents who Epstein was communicating with, and about what. There have been some interesting analyses so far, but I haven’t seen the social networks analyzed over time yet.

I am envisioning a data exploration web app as a first analysis where you can observe timelines of changes in Epstein’s social network, and look at what specific emails said using the interface.

This is an opportunity for me to prototype a component for a new app I’m developing currently called FutureShaper for planning and analyzing social change efforts.

Below is a detailed overview of the Epstein Estate Documents repository intended to help you get started if you want to poke around or analyze the data.

Thanks so much for reading. Please subscribe to stay posted about my latest practical lessons on how to apply social science to your work and life, if you don’t already. Subscribing on Substack helps all of us connect who are working towards improving human cooperation and coordination, and reducing corruption in the democratic process.

Subscribe now

Data Overview

There are four directories in the Documents repository. Here is what I understand about each:

• IMAGES: Contains a set of 12 directories with images (.jpg format), which appear to be mostly scanned documents. The first directory, IMAGES/001 has James Patterson’s and John Connolly’s (2016) book with Tim Malloy, Filthy Rich (see below). This seems to give us some basic facts that we the people can use to orient ourselves.

  • I poked through directory 007: it starts with an email Epstein received on June 30, 2019, just over a month before his death: it’s a summary of news articles from “flipboard.com,” a news aggregator. The first headline is “What Does Putin Really Want?”

  • I clicked forward and there was A Resource Guide to the U.S. Foreign Corrupt Practices Act (FCPA) from 2012.

• NATIVES/001: The NATIVES directory has only one subdirectory, 001. Here there are some interesting spreadsheets about finances and gas usage at Mar-a-Lago. There’s also a video of a dog gnawing on a Donald Trump chew toy that was first uploaded to this repository in April 2020, in .mov format (see first analysis below). The video’s metadata says it was made June 29, 2017.

• TEXT: This directory has communications in .txt format in two directories…

  • 001, which contains stuff like this, some kind of legal letter sent by Kenneth W. Starr to Judge Mark Filip in defense of Epstein on May 27, 2008. Here’s an interesting excerpt: We urgently requested that your Office conduct an independent review of the proposed federal prosecution of our client, Jeffrey Epstein. The dual reasons for our request that you review this matter are (i) the bedrock need for integrity in the enforcement of federal criminal laws, and (ii) the profound questions raised by the unprecedented extension of federal law by the United States Attorney’s Office in Miami (the “USAO”) to a prominent public figure who has close ties to former President Clinton. The need for review is now all the more exigent. On Monday, May 19, 2008, First Assistant Jeffrey Sloman of the USAO responded to an email from Jay Lefkowitz informing U.S. Attorney Alex Acosta that we would be seeking your Office’s review. Mr. Sloman’s letter, which imposed a deadline of June 2, 2008 to comply with all the terms of the current Non-Prosecution Agreement (the “Agreement”)

  • 002 contains mostly the emails—though one of the documents is a reading list with table of contents and several articles, which seem to have been prepared for Epstein (see HOUSE_OVERSIGHT_031913.txt). These can help us understand the social dynamics of Epstein, Trump, Maxwell, and all the other super-frenemies. Here are two examples from correspondence between Epstein and Ehud Barak, former prime minister of Israel—and in this relationship Barak needed Epstein: Epstein was Barak’s fixer, indicating Epstein sat atop Barak in their circle’s hierarchy. See this article by Drop News and recent Democracy Now! episode with Drop News.

    • In 2013, Epstein gives advice that Ehud Barak, former prime minister of Israel, should write an op-ed (HOUSE_OVERSIGHT_031964.txt): I would use the opportunity to compare it with Iran. The solutions become more complex with time, not less. I think many people would like your views on Egypt, Syria, etc. Russia’s role? I think you might point out the gassing of “women and children” is an expression from the 20th century. Women are no longer equivalent to children; civilians vs. combatants only.

    • In 2017, Ehud Barak forwarded Epstein a Twitter post about how the Israeli government was going to make a Trump “welcome video” that would appeal to Trump by showing him “Where God had sex with Maria”

• DATA: these contain relational data and metadata about different images, etc.; they are used in Concordance software, apparently, but I know little about this and I’m not sure how useful they are anyway. I downloaded the files, uploaded them to ChatGPT, which read them. I don’t know how they’re useful yet. If you, or someone you know, knows anything about Concordance, please drop a message here or via direct message! The screenshots below show what ChatGPT read from the files—I just downloaded the two files to my computer from the cloud, then re-uploaded them into ChatGPT.

Note! It was surprisingly difficult to get ChatGPT to give me exactly what I wanted. The following prompts/interactions got me these previews:

This is the script for Episode 4 of my five-part miniseries The Logic of Examined Science, distributed as a YouTube Shorts Playlist (@mat-phd-ss4s)—for first cuts and previews of content like this please follow me on Instagram: @matthewadamturnerphd

The Logic of Examining Science

Hi everyone, so this week we’ve been working through the Logic of Examined Science. Yesterday, we put logic to work to prove that unexamined science isn’t worth doing. Today I want to drive home why we examine science, and how we do it.

Why We Examine Science: To Prune

I have a tomato plant that struggled through the summer but it’s making a comeback. I think I can help it by pruning it a little bit.

Tomatoes need to be pruned. You remove the dead branches so the healthy parts can grow. When we do that we make room for these new shoots and better distribute resources.

So why do we examine science? Because some ideas, methods, and theories need to be trimmed back if they are no longer bearing fruit.

How We Examine Science: Metascience

Now the question is: how do we decide what to prune—how do we go about examining science? The answer is called metascience—the science of science.

One important foundational model of metascience is the Critical Rationalism model, invented by Karl Popper. It explains how science disciplines come and go: Now I’ll use another farm metaphor but sillier this time—let’s say Popper saw Science like a Chicken.

Popper says that science disciplines are like a chicken and egg all together—like sociology or cognitive science, let’s say. If science is working normally, we should always see chickens hatching out of chicken eggs. If we get a surprising result it’s like a dinosaur hatched out of a chicken egg, something that’s treated like it’s nearly impossible.

Feyerabend comes later to explain that Popper misses the fact that new paradigms emerge precisely because they abandoned old orthodoxy and dogma—research can’t be compared as easily as we could discern a dinosaur and a chicken. Science is more like a messy chicken coop that needs to get cleaned up now and then.

These days we have to deal with the fact that several crises are brewing at once in the social sciences. The experimental methods don’t work like people thought; overblown, overgeneralized studies claim to reveal some truth about our world, but too often they actually mislead us.

Modern metascience

Modern metascience is a synthesis that envisions science as a bubbling cultural cauldron—and things aren’t looking great for some flavors of social science and psychology.

In 2005, John Ioannidis used mechanistic modeling to explain that most published research is false, starting from simple assumptions about how frequently psychological effects, for example, are truly activated in experiments—like the group polarization effect which I reviewed in a recent Substack newsletter and in a short lecture available on YouTube.

Paul Smaldino has pioneered the use of cultural evolution to famously explain why bad science “evolves”—culture evolves depending on what behaviors are rewarded, just like biological species evolve to find a niche—the niche of a catfish is to troll the bottom of freshwater lakes and streams, for example.

So scientific crises are caused when scientists, on balance, are rewarded more for a large corpus of sexy, intriguing results, instead of rewarding the overall quality of their work in terms of advancing our understanding of the world.

Conclusion

So to conclude, metascience is how we examine science. We do metascience to help us understand and guide scientific renewal—we remove detritus to make room for growth.

Thanks so much, and please like, subscribe, and especially comment so we can continue our journey together. See you next time.

References

Ioannidis, J. P. A. (2005). Why most published research findings are false. PLoS Medicine, 2(8), 0696–0701. https://doi.org/10.1371/journal.pmed.0020124

Smaldino, P. E. (2023). “The Scientific Process” in Modeling Social Behavior: Mathematical and Agent-based Models of Social Dynamics and Cultural Evolution. Ch. 8. Princeton University Press. [link]