At The Overlap

When Food Debates Flip-Flop

Every few months, a food you thought you understood becomes controversial again. Coffee. Butter. Oat milk (or oat drink). Seed oils. At some point, you start wondering: Is science confused? Can it just decide already?

I’ve lost track of how many times the coffee headline has flipped. One year it raises blood pressure. Another year it extends lifespan. You start thinking: Should I drink it? Avoid it? Measure my consumption to the millilitre?

The coffee didn’t change. The lens did. The study design changed. The population. The research question. Still, when conclusions shift, it can feel like science is changing its mind. And honestly? I used to find that unsettling. Frustrating. Like that friend who reads the entire menu out loud and still says, “I don’t know what to order.” (And by friend I mean me.)

I was always quite good at remembering things in school. Formulas. No problem. Definitions. Also fine. Back when learning meant studying facts for a test. Knowledge felt fixed.

Do you remember when we all learned that fat was bad? Cholesterol was dangerous and salt needed to be reduced. (Or at least that’s how it was communicated.)

It’s not that the biology suddenly changed. Our understanding of it did. And that understanding is always filtered through models, definitions, classifications. Ways of turning messy biology into something we can label, regulate, and argue about.

Somewhere along the way, that stopped frustrating me as much, and started fascinating me more.

---

Key Takeaways

• Food debates often feel chaotic. But that’s not because science is unstable.
• Guidelines and classifications are tools designed to simplify complex realities, like recommendations on which fats one should eat or ultra-processed foods definitions.
• When those tools evolve, it can feel like flip-flopping, especially in public conversation.
• Many arguments aren’t about facts, but about which lens or model we’re using.
• The real tension lies between simplified categories and dynamic systems.

---

When Facts Seem To Change 

It was 2012. I remember sitting in a food law and regulations course. And no, we weren’t memorizing the regulations. We were studying the history of it all across a few different areas. Canada. The USA. The EU. Codex Alimentarius (Latin for ‘Food Code’, which is a collection of internationally recognised food standards, guidelines and other recommendations).

I remember thinking: Wow, that is crazy how much this has changed. Repeatedly.

That course also had my first open-book exam. Why? Because there’s no point memorising details that will change. The real skill is knowing where to look. And how to interpret what you find.

Guidelines shift. Labelling requirements evolve. Definitions get revised. Not because regulators can’t decide, but because these are all tools built to solve specific problems at specific moments in time.

But in public conversation, those shifts rarely feel like careful updates. They feel like flip-flopping.

And that’s when the debates start.

These days food debates are everywhere. Seed oils are toxic. Or they’re perfectly fine. Plant-based meat is the future. Or it’s ultra-processed disaster. Constant headlines declaring entire food groups heroes or villains.

What fascinates me (and at the same time frustrates me) isn’t disagreement. It’s how attached we often get to simplified versions of complex realities — and then treat the simplification as if it were the whole picture.

Where It Gets Really Interesting: Food Itself

Working in food changes how you read headlines. Food isn’t just biological matter. It’s chemistry, physics, processing, regulation, behavior, culture. All layered together.

Take the clean label topic. People want shorter ingredient lists, fewer additives and more “recognizable” names on the food package.

On the surface, that sounds completely reasonable.

You want lecithin removed? Well, it is used both as an emulsifier and surfactant, often contributing to texture and shelf-life of food products. It is often what keeps chocolate smooth and prevents oil separation in spreads and salad dressings.

Mono- and diglycerides act similarly to lecithin. One common use is in packaged bread where they help it stay soft instead of it turning into a brick by day two.

I could go on.

But this isn’t about defending ingredients. It’s about understanding how we think about them.

Remove them, and yes, the label looks cleaner. “Look at my clean label!” (Now ignore the watery chocolate milk.)

Making the ingredient list shorter doesn’t suddenly make the system simpler. Without careful reformulation, the product will noticeably be different, and not in a good way.

A similar dynamic exists with nutrition labels, something I wrote about in my last post. The numbers seem simple at a glance. But are much more complex when you delve deeper.

The more I understand about the complexity of food systems, the more most debates (not just about food) look less like arguments about facts — and more like arguments about which lens we’re using to interpret the same complexity.

Choosing Curiosity, Not Sides

Once I started seeing debates this way, my own questions began to shift. Somewhere along the way, I stopped asking “What’s right?” And started caring a lot more about “How was this conclusion built?”

Things like:

• What assumptions sit behind this classification?
• What problem was this guideline designed to solve?
• In which context is this claim valid and where might it break down?
• What trade-offs are being managed that we don’t immediately see?

In any debate, I’m less interested in taking sides and more interested in understanding how the categories behind those sides were constructed.

Because nuance is usually where reality actually lives.

Why It Might Feel So Intense Right Now

We live in a time that loves clean categories. Health scores. Food rankings. Sustainability ratings. AI-generated answers. Everything looks like a dashboard. Preferably colour-coded.

Ultra-processed foods are one example that dominates headlines right now. The classification captures something real about how modern foods are formulated.

So while calling something “ultra-processed” can feel definitive, it’s a classification model — one way of organising a very complicated food landscape.

The issue is that once a category enters public conversation, it often gains more moral clarity than the underlying model was designed to carry.

Processing becomes equated with harm. “Natural” becomes equated with health. People get caught up in using the term and forget that reality is rarely that binary.

It’s not just ultra-processed foods. It’s carbon labels. Food product sustainability scores. Protein scores. AI apps telling you what to eat based on a photo.

The same thing happens outside food too. Sustainability metrics. Climate models. Even GDP figures.

It’s the same pattern. The number or category looks like a verdict. As if the final word has been spoken.

Underneath it? Assumptions. Definitions. Modelling decisions.

Maybe the Messiness Isn’t a Problem

When a debate about food starts, I often get the feeling that an argument is going to start. Someone wants to prove that one side is the right one. Which makes me think the discomfort isn’t that science changes. Things change all the time.

Maybe it’s that we forget our categories were always simplifications to begin with.

And honestly, the same dynamic shows up anywhere data intersects with personal values.

When we recognise that guidelines, classifications, and numbers are tools — models built to simplify complexity — the messiness starts to make sense.

And at least for me, far more interesting.

That’s the space I find myself pulled towards. Between structured knowledge and lived complexity.

If you’ve ever felt that food debates feel strangely chaotic, you’re not imagining it. It’s not because science is broken. It’s because we’re asking simplified categories to fully represent dynamic systems.

And that tension — between model and reality — is what I’ll keep exploring here At The Overlap.