E. M. Forster · 1879
5–8 unknown words per 100 at CET-4 level. Looking each one up will stall you; this one wants glosses that appear in place.
Coverage = the share of words in the text you already know. The research gives two lines: 98% before independent fluent reading is realistic, 95% for the instructional level that needs support (Hu & Nation, 2000). Proper nouns are not counted as unknown — recognising that it is a name is enough; this book has 598 of them.
CET-6 and up, or read it with glosses. 109,600 words in total; at 150 words a minute for a non-native reader, about 12 h 11 min to finish. Average sentence length 11.2 words — short sentences, brisk pace. Reaching 95% coverage on this book (the floor for readability) takes roughly 16,000 word families.
The opening of Howards End, word for word. Tap any word you don't know and its meaning here appears beside it — no sign-up, tap as often as you like.
It isn’t going to be what we expected. It is old and little, and altogether delightful—red brick. We can scarcely pack in as it is, and the dear knows what will happen when Paul (younger son) arrives tomorrow. From hall you go right or left into dining-room or drawing-room. Hall itself is practically a room. You open another door in it, and there are the stairs going up in a sort of tunnel to the first-floor. Three bedrooms in a row there, and three attics in a row above. That isn’t all the house really, but it’s all that one notices—nine windows as you look up from the front garden.
👆 Tap any word to try it
The numbers above are computed, but we haven't cut Howards End into passages yet.
If you have the ebook, upload it — ready in a few minutes, same three-minute passages, same tap-a-word glosses. Free, no card needed.
These are already built, and closest in difficulty to Howards End.
Ordered by unknown words per 100. Lower reads easier; to step up a level, pick a higher number.
Computed by Readii from the public-domain text (updated 2026-10-21), not estimated by hand. The Readii reading level is our own scale, anchored on unknown-word density; it is not the same system as Lexile or AR and does not convert directly. Unknown words are measured against the CET-4 ∪ Gaokao lists, so this is difficulty for a Chinese-speaking learner, not a native-child reading level.