EFLAW: the readability score built for non-native English readers

You write a sentence you think is simple: "By and large, the upshot is that we ought to take it on board." Every word is short. There isn't a single three-syllable term in there. A classic readability formula like Flesch would rate it as effortless. Yet a reader whose first language isn't English may stall on it completely — because the difficulty isn't in the words, it's in the idioms, the phrasal verbs, and the dense little function words crammed together. This is exactly the gap that the McAlpine EFLAW score was built to measure.

EFLAW stands for "English as a Foreign Language Average Word" length, a readability metric devised by the New Zealand writer Rachel McAlpine specifically for evaluating text aimed at ESL readers and global audiences. If you write documentation, support content, public-sector notices, or anything else that will be read by people across many first languages, EFLAW often tells you something the syllable-counting formulas miss.

Why syllables are the wrong signal for ESL readers

Almost every famous readability formula — Flesch Reading Ease, Flesch–Kincaid, Gunning Fog, SMOG — leans heavily on syllable counts as a proxy for word difficulty. The logic is that longer, polysyllabic words tend to be rarer and harder. For a native reader that mostly holds up.

For a non-native reader it breaks down. Long words in English are frequently Latinate and therefore transparent to speakers of Romance languages — "information", "necessary", "communication" are long but instantly recognisable to a French or Spanish speaker. Meanwhile the genuine obstacles for ESL readers are the short words: the prepositions, particles, and helper verbs that swarm together in idiomatic English. "Get it across", "put up with", "by and large" — none of those words is long, yet the meaning is opaque if you learned English as an adult. A formula that rewards short words actively misleads you here.

How the EFLAW score is calculated

EFLAW turns that insight into a formula. It counts two things: the total number of words in the text, and the number of "mini-words" — words of three letters or fewer. The score is:

EFLAW = (words + mini-words) ÷ sentences

The McAlpine EFLAW calculator implements exactly this: it strips punctuation from each token, counts every word whose length is three characters or less as a mini-word, then divides the sum of total words and mini-words by the number of sentences. The effect is clever. A sentence is penalised both for being long (more words per sentence) and for being clotted with tiny words (more mini-words per sentence). A short clean sentence scores low; a long sentence stuffed with "of the", "to be", "in a", "on it" scores high because each of those little words gets counted a second time.

The result lands on a band scale that the tool reports alongside the raw number:

• 20 or below — very easy. Comfortable for most ESL readers.
• 21 to 24 — easy. Still accessible to a global audience.
• 25 to 28 — moderate. Starting to demand fluent reading.
• 29 and above — difficult. Likely to lose non-native readers.

Because the formula needs no syllable dictionary and no list of "hard" words, it is fast, deterministic, and language-mechanical — there's nothing fuzzy to disagree with. That simplicity is part of why it travels well.

Reading the number in practice

Suppose you run a paragraph and get an EFLAW of 31. The tool also shows the mini-word count, and that's where you look first. A high mini-word ratio is the tell that your sentences are idiomatic and grammatically dense rather than simply long. The fix usually isn't "use shorter words" — they're already short — it's to break the sentence apart and unpack the phrasal verbs. "We need to take on board the points you raised" (EFLAW-unfriendly: lots of mini-words) becomes "We will consider your points" (fewer words, far fewer mini-words, and a plainer verb).

The other lever is sentence count. Because you divide by the number of sentences, splitting one sprawling sentence into two well-formed ones lowers the score directly. EFLAW rewards exactly the habit that helps ESL readers most: one idea per sentence.

When to reach for EFLAW — and when not to

EFLAW is the right tool when your audience is genuinely international or non-native: global product documentation, multilingual customer support, instructions, health and safety information, or government communication that has to reach everyone. In those contexts it correlates with real-world comprehension better than the syllable formulas, which were validated on native-speaking American schoolchildren.

It is the wrong tool when you're writing for a native, specialist audience where long technical vocabulary is fine and expected — a research abstract or a legal brief. There, EFLAW's blindness to word complexity becomes a liability: it will happily pass a paragraph full of jargon as long as the sentences are short. For that audience, pair it with a syllable-based formula or a vocabulary measure. As with every readability metric, treat the score as a flag, not a verdict: it tells you which sentences to look at, and your judgement does the rest.

If you write anything that crosses a language border, paste it into the EFLAW readability calculator, watch the mini-word count as much as the score, and rewrite the sentences that climb above the "easy" band. It's a small habit that makes your writing genuinely more welcoming to the large share of your readers who are meeting English as a second, third, or fourth language.