Signal and Noise

Chakiaki
4 days ago
9 min read

What Intelligence Analysis Taught Me About Truth—and Why Most People Are Looking for It Wrong

Chakiaki | Chakiaki LLC

There is a particular kind of quiet that comes over you the first time you realize the answer was in front of you the whole time. Not hidden. Not encrypted. Just buried under so much ordinary traffic that nobody bothered to look.

That was most of the job.

As a 1N451 Network Intelligence Analyst for the Air Force, I was not trained to be clever. Clever gets people killed. I was trained to be disciplined—to hold a question steady long enough for the answer to surface on its own and to distrust the answer that arrived too fast because it flattered what I already believed.

I want to talk about that discipline, because I think it is the thing our culture has lost, and I think losing it explains almost everything else.

The Volume Problem Is Not the Problem

Everyone knows we are drowning in information. That observation is so common it has become its own kind of noise.

But volume alone was never the hard part. Volume is a resource problem—more storage, more processing, more eyes. The hard part is that noise is not random. Noise is shaped. It has structure, momentum, and increasingly, intent.

This is not a suspicion. It is documented. Researchers who built a training tool around the actual production of misinformation identified a repeatable set of techniques used to manufacture it: polarization, invoking emotion, conspiratorial framing, trolling, deflecting blame, and impersonating credible accounts (Roozenbeek & van der Linden, 2019). Those are not accidents of a crowded internet. They are methods. Somebody learned them, and somebody applies them on purpose.

Here is what that means practically. When you sit down to figure out what is true about anything that matters—an election, an economy, a war, a man's reputation—you are not sorting truth from static. You are sorting truth from carefully manufactured counterfeits of truth, built to pass the exact checks you were taught to run.

That changes the work. Verification alone stops being sufficient, because the counterfeit was engineered to survive verification. You have to move upstream and ask a harder question: who benefits if I believe this, and what did it cost them to put it in front of me?

There is good news buried in the same research. Because the techniques are finite and documented, they can be taught — and in a study of roughly 15,000 participants, people who were walked through how misinformation gets built became measurably better at spotting it afterward, regardless of their education level, age, or politics (Roozenbeek & van der Linden, 2019). Discernment is not a gift. It is a trainable skill, and the training is not long.

Three Habits That Survive Contact With Reality

I am not going to pretend there is a system. There isn't. But there are habits, and they hold up outside a secure facility exactly as well as they held up inside one.

1. Separate the Observation From the Interpretation

This is the discipline people skip most and pay for hardest.

A video is an observation. What the video means is an interpretation. A job's number is an observation. What that number says about your country is an interpretation. The two arrive fused together—the interpretation delivered in the same breath as the fact, by the same voice, in the same confident tone—and pulling them apart is entirely on you.

When you do, you will notice something uncomfortable: the observation is frequently solid and the interpretation is frequently garbage. Most people then reject the whole package once the interpretation collapses and lose a real fact in the process. That is a defeat you inflicted on yourself.

2. Weight the Source by Its Track Record, Not Its Credentials

Let me use myself as the example, since it costs me something to do it.

I hold a master's degree in cybersecurity and a stack of industry certifications. Those are real, I earned them, and they are close to the least interesting thing about whether I can tell you what is true. What they prove is that I could pass a test written by someone else about a body of knowledge someone else selected. That is not nothing. It is also not the same skill.

Credentials tell you what a person was permitted to learn. Track record tells you what they actually got right when being right was expensive. Those two correlate far less than you would hope, and the gap between them is where most confident wrongness in public life lives.

The uncomfortable corollary is that you have to keep score on yourself the same way. Most people cannot stand to do that, which is precisely why doing it is an advantage.

3. Assume You Are the Target

Not in a paranoid way. In an operational way.

If you can be moved, someone has already calculated what moves you—and the emotions exploited first are the honorable ones. Loyalty. Outrage at injustice. Love of your people. Protectiveness over your children.

That is not an accident, and it is measurable. In an analysis of over half a million social media messages about contested political issues, each additional moral-emotional word in a message increased its spread by roughly 20%—and that amplification occurred largely within ideological groups rather than across them (Brady et al., 2017). Moral language does not persuade opponents. It energizes your own side and travels fast doing it.

Read that again with an operator's eye. If you wanted to move a population, you would not use cheap emotions; they make unreliable levers. You would use the noble ones, because they are dependable, and you would aim them inside a group that already trusts each other.

So the moment a piece of content makes you feel righteous, that is the moment to slow down. Not to disbelieve it. To slow down.

Why I Care About This More Now, Not Less

I spent my career inside institutions built to do exactly this work—Air Force intelligence, then the civilian side of the Department of Defense. Then I chose to leave.

There was a day it stopped being abstract. The specifics stay out of this, because the specifics are the least important part. The pattern is what matters, and you will recognize it.

There was a problem. A real one, with a real cause, and the cause was structural—the kind of thing that takes work, money, and somebody's accountability to fix. That fix did not happen. Instead, the organization located a person.

He became the explanation. Difficult. Hard to have a conversation with. Not a team player. Once that label was attached and accepted, the underlying problem no longer required solving, because it was no longer a problem—it was a personality. Management had produced an answer that cost them nothing.

This has a name in the safety literature, and understanding it changed how I read every institution I have dealt with since. Reason (2000) distinguished the person approach to failure from the system approach. The person approach treats failures as arising from individual carelessness, negligence, or moral weakness, and responds with disciplinary measures, retraining, and what Reason bluntly called naming, blaming, and shaming. The system approach examines the conditions people were working under.

Reason's key insight is the one that applies far beyond hospitals and aviation: the person approach is attractive to organizations precisely because isolating a blameworthy individual severs the failure from institutional responsibility. It is not merely a mistake in reasoning. It is a mistake that pays.

So I looked at what actually happened, and the account did not hold. And I said so. Not as a protest, not loudly—as the answer to a question I had been asked, which is the only thing I was ever trained to give.

Nobody showed me where my read was wrong. Nobody walked me back through evidence I had missed. I was called out and I was attacked. Not for being inaccurate. For not siding with the room.

That is the tell, and I want you to hold onto it, because it applies far outside a government office.

When an organization is healthy, a wrong assessment gets corrected. When it is not, an inconvenient assessment gets punished.

Those are entirely different responses, and which one you get tells you precisely what that institution is actually optimizing for. Most people only run that test once. Most do not recognize the result when it comes back.

And the consequences are documented. A systematic review of employee voice and silence found that concerns raised by staff are frequently not acted upon until a disaster point is reached, that fear of retaliation and job loss are among the primary reasons people withhold information, and that formal "speak up" interventions produce disappointing outcomes when the surrounding professional culture does not actually support them (Lainidi et al., 2023). Read that carefully. The problem is not that organizations lack reporting channels. Most have them. The problem is that channels do not overrule culture.

There is a companion tell worth learning alongside the first: when a systemic problem gets converted into a personality problem, the organization has stopped trying to solve it. Watch for the moment a structural failure acquires a name and a face. That is not an investigation reaching its conclusion. That is an investigation being closed early, at somebody else's expense.

What I understood afterward was that the day itself was not the change. The day was only when I finally saw it. The real shift had been underway for years, in a direction nobody names out loud: the operative question drifting from what is true toward what is acceptable to say is true. Those are different questions, and given enough time they produce different answers. Nobody announces that drift. Nobody sends that memo. It just becomes gradually easier to agree and gradually more expensive to check, in increments small enough that objecting to any single one would make you look unreasonable.

I had been walking down that slope for a while. The incident was just the moment I looked up and saw how far down I already was. Given the choice between truth and compliance, I picked truth—and then I found out what that costs, which is the part nobody covers in the training.

Why a Song and Not a Report

I write and record now. People assume that means I set the analytical work down. It means the opposite.

Take Greenwood. The Greenwood District of Tulsa, Oklahoma, was a self-sustaining Black commercial center—banks, hotels, theaters, and physicians—known as Black Wall Street. On May 31 and June 1, 1921, a white mob destroyed it. This is not contested history recovered by amateurs. The State of Oklahoma convened a commission that spent four years examining it, and its final report documents the destruction, the involvement of local authorities, and the subsequent suppression of the record (Oklahoma Commission to Study the Tulsa Race Riot of 1921, 2001).

That report has been publicly available and free to download since 2001. The facts were never hidden. They were simply not carried—largely absent from the curriculum I grew up with, and largely absent from most people's.

You cannot argue somebody into caring about a fact like that. I have watched people be handed it and set it down.

So I researched it, and then I built it into a track. Three minutes. People who would never open a commission report have now heard it, and some went looking afterward.

I want to be careful here, because this essay would be a hypocrite otherwise. The claim that music makes material easier to learn is more contested than artists like me tend to admit; researchers comparing sung, spoken, and written presentations of the same text have found the effects depend on whether you are measuring recall or comprehension, and the advantage does not run in one simple direction (Lehmann & Seufert, 2018). So I will not tell you a song teaches better than a book. I do not think the evidence supports that cleanly.

What I will tell you is narrower and, I think, defensible: a song gets encountered by people a report never reaches, and it survives in circulation long after the report is archived. Distribution is not the same as pedagogy. But a truth nobody encounters teaches nobody anything.

The album that grew out of that work, The Reconstruction, is documented like a film—over 150 sources behind twelve songs. Not because anyone asked for footnotes on an album, but because if I am going to stand in front of people and say this happened, I had better be able to show my work when someone checks. The delivery format changed. The standard did not.

What This Asks of You

Self-reliance gets talked about mostly in terms of supplies and skills. Food, water, a generator, and a trade. All real. All necessary.

But the first system you have to get off the grid is your own judgment. If you cannot determine what is true without being told, then everything downstream — your vote, your money, your children's education, what you are willing to fight for — is being operated by somebody else, no matter how much rice is in your garage.

The research says this capability is trainable (Roozenbeek & van der Linden, 2019). Nobody is going to hand it to you. It gets built the way any capability gets built: badly at first, repeatedly, with honest scoring of your own failures.

Start there. Everything else follows.

References

Brady, W. J., Wills, J. A., Jost, J. T., Tucker, J. A., & Van Bavel, J. J. (2017). Emotion shapes the diffusion of moralized content in social networks. Proceedings of the National Academy of Sciences, 114(28), 7313–7318. https://doi.org/10.1073/pnas.1618923114

Lainidi, O., Jendeby, M. K., Montgomery, A., Mouratidis, C., Paitaridou, K., Cook, C., Johnson, J., & Karakasidou, E. (2023). An integrative systematic review of employee silence and voice in healthcare: What are we really measuring? Frontiers in Psychiatry, 14, 1111579. https://doi.org/10.3389/fpsyt.2023.1111579

Lehmann, J. A. M., & Seufert, T. (2018). Can music foster learning—Effects of different text modalities on learning and information retrieval. Frontiers in Psychology, 8, 2305. https://doi.org/10.3389/fpsyg.2017.02305

Oklahoma Commission to Study the Tulsa Race Riot of 1921. (2001). Tulsa Race Riot: A report by the Oklahoma Commission to Study the Tulsa Race Riot of 1921. Oklahoma Historical Society. https://www.okhistory.org/research/forms/freport.pdf

Reason, J. (2000). Human error: Models and management. BMJ, 320(7237), 768–770. https://doi.org/10.1136/bmj.320.7237.768

Roozenbeek, J., & van der Linden, S. (2019). Fake news game confers psychological resistance against online misinformation. Palgrave Communications, 5, 65. https://doi.org/10.1057/s41599-019-0279-9

Every source above is free to read in full. No paywall, no login, no institutional access, no request form. Check them.

Chakiaki is an Air Force and Army veteran, former intelligence analyst, and independent recording artist. Research facts. Present truth. Educate. Reclaim self-reliance.