Word Bocce

How to play

1. Your red ball starts on one word. The yellow jack is another word. Roll your ball as close to the jack as you can.
2. Tap words to add (+) or subtract (−) their meaning, up to three, then tap Throw.
   hat + foot − head → shoe
3. Closeness is a ranking. 1st means the jack is the nearest word to your ball, a bacio (a kiss). You get four balls; your best counts.

Play the 30-second tutorial

Why does adding words work?

The game stores each word as a long list of numbers: its position on a map of meaning. Words with similar meanings get similar lists. A throw adds and subtracts those lists, so the ball moves through meaning.

You can choose which map the game uses: common sense, or raw text.

Why don't throws always make sense?

After any throw, tap Why? for a reading of what each word did. It's a best guess: why the game can't fully explain itself.

Versus rules

Real bocce rules: whoever is farther from the jack throws next. When both sides are out of balls, the closer side scores a point for each of its balls that beats the other side's best.

Two maps of meaning

Every word in the game has a position on a map of meaning: a long list of numbers. Words with nearby positions count as close. You can choose where those positions come from.

Common sense

Built from how words are used in text, plus ConceptNet, a large collection of everyday facts that people have written down: "a hen lays eggs", "bacon is a kind of meat", "you wear a shoe on your foot". Throws mostly work the way you'd expect. It uses about 21,000 everyday words, and a court is only dealt if its best throw can be explained word by word.

Use common-sense words

Raw text

Learned only from which words appear together in billions of words of Wikipedia and news. This is the classic "word embedding" (GloVe) behind the famous king − man + woman ≈ queen. Nobody told it any facts, so it can surprise you: "ham" sits among football clubs (West Ham), and "eggs" sits nearer "meat" than "bacon", because that's how the news writes about them. Try it if you're curious how a machine that only reads text arranges words. (Why is that so hard to explain?)

Use raw-text words

Each has its own Daily. In an online room, the host's choice applies to everyone.

Why did that happen?

Why the game can't fully explain itself

The "Why?" notes are our best reading of the numbers, not the real reasons. A few things make this hard:

• Meaning is spread across hundreds of numbers. Each word's position is a list of 100–300 numbers learned from data. No single number means "food" or "meat"; every number carries a bit of everything.
• We can measure closeness exactly, but not why. Two words end up close because of patterns in what the model learned from. Some patterns are ones people wouldn't expect: surnames (Bacon), sports clubs (West Ham), or which things news stories happen to mention together.
• One position holds every meaning of a word. "Bacon" is a food and a surname; "ham" is a meat and a football club. Subtracting one sense can leave a stranger one behind.
• Language isn't tidy either. Bacon and eggs go together at breakfast, but eggs also belong with nests, hens and farm produce. People disagree about what's "close" too.
• Famous examples are the tidy ones. "king − man + woman ≈ queen" works, but many analogies like it don't, and some published ones were helped along by quietly ruling out the obvious answer (Nissim, van Noord & van der Goot, 2020).

Modern AI systems, including chatbots, are built from learned numbers like these at a vastly larger scale. Working out what they represent and why they behave as they do is called interpretability. It's an open research problem: researchers can now find some meaningful features inside models, but nobody can fully explain a large one.

Further reading: Mapping the mind of a large language model (Anthropic, 2024) · Zoom In: an introduction to circuits (Olah et al., Distill, 2020).