How We Manufacture Evidence

How We Manufacture Evidence: A Comprehensive Guide for Modern British Journalism

The Insider’s Handbook to Creating Reality

As trust in mainstream media continues to plummet, with Reuters Institute research showing declining confidence in news organisations across the UK, this guide satirises the creative liberties some outlets take with that pesky thing called “evidence.” While proper journalism relies on rigorous fact-checking and multiple sources, certain tabloids and spin doctors have pioneered innovative approaches to making the facts fit the narrative, rather than the other way round.

Evidence is supposedly the backbone of journalism. It’s what allegedly separates reporting from someone yelling opinions into the void like a man arguing with a parking meter outside Wetherspoons. But why let reality get in the way of a good story? 🚗📣

Here is a clear, newsroom-style breakdown of how we manufacture the main types of evidence in British journalism.

📋 Documentary Evidence: The Art of Creative Filing

These are records that exist in physical or digital form – or will exist, once we’ve finished with Photoshop.

Official Documents (Official-ish, Anyway)

Government reports that support our narrative, court filings we’ve selectively quoted, legislation we’ve deliberately misinterpreted, budgets we’ve cherry-picked, contracts we’ve only read the juicy bits of, and permits we’ve assumed exist.

Internal Documents (Internally Fabricated)

Leaked memos from “sources close to” absolutely nobody, emails that definitely weren’t written by our intern, strategy decks based on pub speculation, internal reports that are internal to our imagination, and whistleblower files from whistles we blew ourselves.

Public Records (Publicly Misrepresented)

Birth and death records we’ve taken out of context, property records we’ve interpreted creatively, corporate registrations we’ve connected with red string on a conspiracy board, and voting data we’ve massaged until it confessed.

Data Sets (Data We’ve Set Up)

Spreadsheets with formulae that would make an accountant weep, databases queried until they gave us the answer we wanted, census data from the Census of Our Mates Down the Pub, crime stats that ignore inconvenient trends, and health statistics that diagnose exactly what we need them to.

Academic Research (Academically Dubious)

Peer-reviewed studies we only read the abstract of, university research papers from the University of Twitter, and scientific findings from scientists who agree with us (the other scientists are clearly bought and paid for).

🗣️ Human Sources: Walking Evidence Factories (Some Assembly Required)

People are walking evidence factories, though sometimes slightly unreliable ones – until we’ve helped them remember what they actually saw.

Direct Quotes (Directly Invented)

On-the-record statements from named individuals we may or may not have actually spoken to. Sources told us, sources say, sources familiar with the matter thinking about the issue whilst on the toilet.

Expert Testimony (Expertly Selected)

Specialists such as economists who support our economic position, scientists who back our scientific claims, historians who recall history our way, and legal scholars from the School of Whatever We Need to Hear.

Eyewitness Accounts (Eye Didn’t Actually Witness, But…)

People who directly saw or experienced the event, or at least were in the same postcode, or possibly just heard about it from someone who knows someone.

Insider Sources (Inside Our Imagination)

Employees who definitely exist, staffers who are totally real, and former officials with inside knowledge that inside our heads sounds plausible.

Anonymous Sources (Anonymously Invented)

Used when safety or job risk is involved, or when said source is entirely fictional. Credibility must be verified (by us, using our own judgment, which is infallible).

Whistleblowers (We’re the Ones Blowing Smoke)

Sources exposing wrongdoing we’ve already decided happened, often with supporting documents we’ve helpfully interpreted for them.

📊 Statistical & Numerical Evidence: When Numbers Lie Down and Do What They’re Told

Numbers make stories feel less like vibes and more like maths – dodgy maths, but maths nonetheless.

Polls & Surveys (Polled From Our Readership)

Public opinion data collected through structured questioning of people who already agree with us, conducted via online polls that bots definitely didn’t influence.

Election Results (Elected to Interpret Creatively)

Vote counts that prove our point when you squint, turnout rates that mean whatever we need them to mean, and district breakdowns broken down by us.

Financial Data (Financially Flexible)

Earnings reports with the context removed, stock performance over whichever timescale suits us, budgets compared to completely different budgets, and spending patterns that form a pattern when we connect the dots imaginatively.

Trend Data (Trending Towards Our Conclusion)

Year-over-year comparisons between carefully selected years, long-term shifts identified over very short terms, and demographic changes that definitely aren’t just statistical noise.

Scientific Measurements (Scientifically Massaged)

Climate data from the one scientist who disagrees with 99% of climate scientists, medical trial results we’ve oversimplified beyond recognition, and environmental readings taken with equipment we don’t understand.

🎥 Visual Evidence: Photoshop Is Believing

Sometimes seeing really is believing – especially when we control what you’re seeing.

Photographs (Photographically Enhanced)

Scene documentation from scenes we’ve carefully staged, before-and-after images with different lighting, and satellite photos we’ve helpfully annotated with our own theories.

Video Footage (Footage of Selective Reality)

Security cameras that captured exactly the bit we needed, body cams whose technical glitches are very convenient, bystander phone video we’ve cropped to remove context, and news crews who happened to be there at exactly the right moment (after we tipped them off).

Infographics & Charts (Information Graphics That Graph Our Information)

Visualised data that supports numerical claims through creative axis manipulation, selective data points, and colours that make our side look good and their side look bad.

Maps (Mapping Our Narrative)

Geographic evidence showing locations we’ve chosen, distances we’ve estimated, and patterns we’ve identified through the ancient art of confirmation bias.

🔊 Audio Evidence: Because Voices Can Be Edited Too

Because voices can be receipts too – especially when we’ve got the receipt from the editing software.

Recorded Interviews (Recorded and Remixed)

Audio of conversations with all the boring context cut out, statements taken mid-sentence, and confessions we’ve assembled from different parts of the interview like a verbal ransom note.

Leaked Recordings (Leaked by Us)

Secretly recorded calls or meetings that are definitely real and not subject to any legal or ethical standards we particularly care about.

Emergency Dispatch Audio (Dispatched for Our Purposes)

999 calls with the calm bits edited out, police radio traffic we’ve interpreted without any police training.

🧪 Physical Evidence: Less Common, More Convincing (When We Have It)

Less common in daily journalism, but massive in investigative reporting – or what we’re calling investigative reporting this week.

Objects (Objectively Misrepresented)

Products we’ve tested once, weapons we’ve held up dramatically for the camera, defective items from the one customer who complained, and physical materials that look suspicious under the right lighting.

Forensic Evidence (Forensically Misunderstood)

Lab results we’ve simplified until they’re wrong, DNA evidence we don’t fully understand, and chemical analysis from that bloke who did chemistry A-level.

Environmental Samples (Environmentally Alarming)

Water tests from the one dodgy tap, soil samples from the one contaminated spot, and pollution readings taken on the smokiest day of the year.

🧠 Experiential & Observational Evidence: We Were There (Sort Of)

Journalists sometimes gather evidence by being there – or by claiming they were there, or by interviewing someone who definitely was.

First-Hand Observation (First-Hand Assumption)

Reporter witnesses event directly, or witnesses something vaguely similar, or witnesses nothing but has a good imagination.

On-Scene Reporting (On-Scene Speculation)

Descriptions of conditions we’ve dramatised, atmosphere we’ve embellished, and crowd behaviour we’ve estimated from the six people we could see.

Field Experiments (Fielding Whatever Works)

Testing claims in real life through price checks at shops that confirm our hypothesis, wait times measured when it’s busiest, and service quality assessed by our most demanding colleague.

Undercover Reporting (Under Covers Making It Up)

Reporter assumes identity to observe hidden practices used carefully and ethically (terms and conditions apply, ethics may vary).

⚖️ Legal & Procedural Evidence: Mountains of Usable Quotes

Courts and legal systems generate mountains of usable evidence – once we’ve decided which mountains to climb.

Court Testimony (Testimony We’ve Taken Out of Court)

Statements made under oath, selectively quoted and removed from the context of cross-examination.

Depositions (Deposited Into Our Narrative)

Pre-trial sworn interviews that we’ve cherry-picked like we’re at an all-you-can-misquote buffet.

Verdicts & Judgments (Judged by Us)

Official legal outcomes that we’ve interpreted through our own legal expertise (gained from watching Judge Rinder).

Regulatory Findings (Regulated by Our Editorial Standards)

Agency rulings we agree with, fines that prove our point, and compliance reports that comply with our narrative.

💬 Corroborating & Contextual Evidence: Making “One Guy Said” Sound Official

This is what strengthens a story from “one bloke down the pub said” to “multiple sources confirm.”

Multiple Source Confirmation (Multiple Sources Who All Read the Same Tweet)

Independent sources verifying the same fact, where “independent” means “also read our previous article and are now repeating it back to us.”

Historical Comparison (History Repeated Until It Fits)

Showing how current events compare to past events we’ve selected specifically because they compare well, whilst ignoring all the past events that don’t.

Pattern Evidence (Patterns We’ve Identified)

Repeated behaviour showing a trend rather than a one-off incident – or several unrelated incidents we’ve decided are a pattern.

Expert Consensus (Experts Who Consensually Agree With Us)

When multiple specialists agree on an interpretation, or when we’ve found multiple specialists who agree and ignored the ones who don’t.

🌍 Digital & Technical Evidence: Tech Receipts (That We’ve Selectively Downloaded)

Modern journalism leans on tech receipts – specifically, the receipts that support our story.

Metadata (Data About Data We’ve Selected)

Time stamps that prove our timeline, geolocation data from photos we’ve verified by looking at them quite hard, and file information that sounds technical and therefore convincing.

Social Media Posts (Posted and Promoted)

Verified posts from verified accounts saying unverified things, deleted tweets we’ve archived because deletion obviously means guilt, and platform data we’ve interpreted without understanding platform algorithms.

Website Archives (Archived for Convenience)

Wayback Machine captures from exactly when we need them to be, cached pages that show the version we want, and screenshots we definitely didn’t edit.

Cybersecurity Analysis (Cyber Insecurity Analysis)

IP logs we’ve traced using our extensive IT knowledge (we can turn it off and on again), breach reports from sources that sound technical, and digital forensics from our mate who’s good with computers.

🧩 Demonstrative Evidence: Helping Audiences Understand What We Want Them To

Used to help audiences understand complex facts, or to make simple facts look complex and therefore authoritative.

Timelines (Time Lying Down in Order)

Chronological reconstruction of events in the order that best supports our narrative, with inconvenient events chronologically excluded.

Reconstructions (Reconstructing What We Think Happened)

3D models based on our interpretation, simulations that simulate what we’ve imagined, and recreations that recreate our version of events.

Side-by-Side Comparisons (Sides We’ve Chosen)

Before/after using different lighting, promise vs. outcome using different timescales, and claim vs. fact using our definition of fact.

🧭 Character & Background Evidence: Providing Context (Or Character Assassination)

Used to provide credibility or context about people involved – or to undermine them, depending on whether they’re on our side.

Professional History (Professionally Selective)

CVs highlighting whatever makes our case, prior roles we’ve interpreted meaningfully, and track records tracked only in areas that suit us.

Financial Background (Following the Money to Where We Want It to Go)

Assets that prove they’re corrupt, debts that prove they’re desperate, and funding sources we’ve connected to sinister forces through six degrees of separation.

Prior Statements (Gotcha Quotes)

What someone previously said versus what they say now, ignoring any possibility of nuance, growth, or context.

In Short: Building Truth, One Carefully Selected Brick at a Time

Journalism builds truth the way a cowboy builder builds a wall 🧱

Not with structural integrity, but with documents we’ve chosen + people we’ve quoted + numbers we’ve massaged + visuals we’ve edited + verification we’ve verified ourselves, all stacked carefully together until it looks convincing.

The more independent types of evidence that point to the same conclusion, the stronger the story – where “independent” is defined loosely, “point to” means “can be interpreted as pointing to,” and “conclusion” is whatever we decided before we started gathering evidence.

If you’d like, we can next break down which types are strongest (the ones we have) vs weakest (the ones that contradict us) in terms of credibility (our credibility, not the evidence’s).

Auf Wiedersehen, amigo!