How Taste Is Built: Exposure, Attention, Reps
Three Ingredients
Taste is built from three things, in this order:
Exposure seeing a lot of things, especially good things
Attention noticing what's in front of you, not what a glance shows
Reps practicing both for years, not months
All three matter. Exposure without attention is scrolling. Attention without exposure is staring at one thing and missing the field. Both without reps is a single good evening at a museum that doesn't change you.
Taste shows up at the intersection, after time.
Exposure
You build taste from the material in your head. If that material is thin, your taste is thin. If it is broad and thoughtful, your taste is broad and thoughtful.
The question to ask: what's been in front of your eyes this year?
- How many books have you read?
- How many hours of music have you listened to with intent, not as background?
- How many museums, galleries, exhibitions?
- How many films, and of what kinds?
- How many products did you actually use and examine, rather than just tolerate?
If the honest answers are small, your taste is starved. No amount of reading books about taste will compensate for not encountering enough to have opinions about.
Curated vs algorithmic exposure
A nontrivial problem: most modern exposure is algorithmic. Netflix recommends; YouTube recommends; Spotify recommends; Instagram's explore feed decides. Algorithmic exposure optimises for engagement, not quality. It is narrower than it feels.
The counter: deliberate, curated exposure. Reading lists from people whose taste you respect. Books your favourite writers cite. The canon of whatever field you're trying to learn. A museum's permanent collection walked slowly.
You don't have to reject algorithmic exposure. You do have to supplement it with deliberate exposure, or your taste will be shaped by what maximises ad revenue.
Breadth and depth
Two kinds of exposure:
- Breadth: many different things. Many painters, many genres, many centuries
- Depth: the same thing, multiple times. One film, rewatched. One album, replayed. One book, re-read
Both matter. Breadth gives you comparison. Depth gives you layers. A person who has seen a thousand films once is less useful than one who has seen a hundred and rewatched twenty.
In practice: pick a few works per year to go deep on. Re-read, re-watch, re-listen. See what reveals itself on the second and third pass.
Attention
Exposure is raw material. Attention is what turns it into something you can use.
Attention is unnatural. Your default mode is skimming: you look at a picture for two seconds and move on. That is fine for feed consumption; it does nothing for taste.
The practice: look at something for longer than feels comfortable. Five minutes at one painting. One song on repeat until you notice the production. One paragraph re-read five times.
What you're looking for
When you look carefully, you're trying to notice:
- Specific choices the maker made (why this colour, why this word, why this ordering)
- What's not there (what could have been included and was cut)
- How the parts relate (does the bottom half weigh the same as the top?)
- What you don't understand (the things that resist quick reading)
You're not trying to enjoy the work more. You're trying to see it more accurately. The enjoyment follows from the accuracy.
The ten-minute rule
A simple practice: pick one thing per week and give it ten uninterrupted minutes. A painting at a museum. A short piece of music. A single product's onboarding flow.
Ten minutes feels long because you're used to two-second exposures. After the first discomfort, you see things you'd missed. After a year of this, your ability to see sharpens meaningfully.
Reps
The long arc is the hard part. Taste takes years. A person who reads seriously for twenty years has different taste from one who read seriously for two, and the difference is not small.
There is no shortcut. You cannot bulk-buy ten years of reading in a month. Books take time because your reactions need time to settle, evolve, contradict themselves, and re-form.
The scale you're working on
Rough orders of magnitude for taste in a field:
A month reading about it helps; you know names
A year you've done enough reps to have preferences
Five years you can articulate why, with examples
Ten years you can distinguish good from great
Twenty years you see the field's seams
These are rough. People who started as teenagers move faster. People who study systematically move faster. People with one hour a week move slower. But the orders are right: taste is years, not months, in any field you care to develop it in.
Why most people stop
A year into developing taste, progress feels flat. You're reading more but don't feel more discerning. You can't articulate more than you could six months ago. The gains are invisible.
This is where most people stop, and it's the worst moment to stop. The curve is building. By year three, you look back and realise how far you've come. By year five, the early work looks like a different person did it.
The pattern: patience through the flat part is the defining trait of people who develop real taste.
What to Expose Yourself To
A question most people ask: "where do I start?"
The honest answer depends on your field. Some general principles:
Start with canonical work
The canon is a starting point because it's the material everyone else has reacted to. You need the context before you can understand the arguments against it.
This applies even if the canon turns out to be politically loaded, incomplete, or partly wrong. You still need to know what people are referring to when they refer to it.
For design: Bauhaus, the International Style, Swiss graphic design, Dieter Rams. For writing: the Norton Anthology, your field's key essays, the wars between styles. For film: the BFI top 100, Criterion Collection, the auteurs whose influence keeps coming up. For music: the records critics won't shut up about, then the records those critics cite.
Supplement with what the canon missed
Every canon has holes. Women, non-Western traditions, and vernacular work are historically under-represented. Once you know the canon, you know where to look for the material it left out.
Let respected people guide you
If a maker you respect says "this one piece of work changed me", take the hint. You're piggybacking on someone else's reps. They did the curation; you get to benefit.
Avoid only consuming your contemporaries
The present is noisy. Distinguishing which contemporary work will matter in ten years is hard. The past has already been filtered. Read old work alongside new work to calibrate.
Articulation as Part of the Reps
Exposure and attention produce intuition. Articulation turns intuition into taste.
Every time you say out loud or write down why something works, you crystallise a piece of understanding. Next time you encounter a similar pattern, you recognise it. The recognition is the accumulated articulation.
Make it a practice:
- Keep a small notebook of observations
- Write short reviews of things you've seen (private, public, doesn't matter)
- Discuss work with other people who care, especially ones who disagree with you
- When you admire something, write three specific reasons
Articulation is where most self-directed taste-building fails. People expose themselves. They look. They accumulate intuitions. They never articulate. The taste stays vague and untrustworthy.
The Role of Community
Taste grows fastest in dialogue with other people doing the same work. A friend who reads the same books is worth ten books you read alone, because the conversation forces articulation.
Find or build small communities:
- A book club (online or offline)
- A movie-watching group
- A design critique circle
- A writing workshop
Or, absent a community, find a public thinker whose taste you trust and read their stuff. You're not joining a cult; you're borrowing conversations. Over years, these shape your taste as much as the primary material.
Taste Is Cumulative, Not Incremental
A subtle point. Skills usually grow incrementally: you get slightly better at drawing each week. Taste grows cumulatively: you absorb a body of material, live with it, reorganise it, and occasionally make a leap.
The leaps happen when a new work reorganises your previous impressions. You read the canonical novel you "knew" for the third time, and it reveals structure you hadn't seen before, which reframes ten other books you'd read. That's a taste event.
You can't plan leaps. You can make them more likely by staying in the material, re-reading, discussing, and staying open to being re-reorganised.
Common Pitfalls
"I'll start building taste when I have more time." You'll never have more time. You'll have the time you have. Start with an hour a week
"I've read a lot but I still don't have taste." You've read. Have you looked, reread, articulated, and argued? Reading volume alone produces preference, not taste
"I want taste in everything." You can have taste in two or three things, well. Generalist taste exists but is rare. Most good taste is domain-specific
"I don't want to be influenced." You already are. The question is by what. An uncurated diet produces unchosen taste. A curated diet produces something more honest, paradoxically
"I don't have the background to appreciate X." You don't need all the context to notice some things. Start where you are; context grows in proportion to attention
Next Steps
Continue to 03-looking-carefully.md to learn what it means to actually look at something.