The applied side — acquisition methods, comprehensible input, and translation as a set of choices, not one answer.
9 topics · 14 curated works
No prior grounding assumed.
On Translating Homer
Matthew Arnold · 1861
Argues a translator of Homer must be judged not against the target language's literary taste but against how a competent scholar hears the original,…
+2 more at this level
Assumes you know the vocabulary.
Preface to Ovid's Epistles
John Dryden · 1680
Divides translation into metaphrase, paraphrase and imitation, and argues paraphrase, keeping the sense but not the words, is the right course…
+5 more at this level
Primary sources and full treatments.
The Art of Teaching and Studying Languages
François Gouin · 1880
Argues from his own failed attempt to learn German by memorising a grammar that language is instead learned by narrating connected series of everyday…
+4 more at this level
12 of 14 works
Matthew Arnold
Argues a translator of Homer must be judged not against the target language's literary taste but against how a competent scholar hears the original, and that Homer's rapidity, plainness and nobility have each been sacrificed by past translators in turn.
link checked 17 Sept 2026Harold E. Palmer
Argues effective language study balances conscious study with unconscious habit-forming practice, and sets out the psychological conditions, such as spaced repetition and graded difficulty, under which each works best.
about 3 hours to readlink checked 17 Sept 2026Stephen D. Krashen
Argues a second language is acquired only from input a learner can understand slightly above their current level, and that conscious grammar study helps monitor speech but does not itself build competence.
link checked 17 Sept 2026John Dryden
Divides translation into metaphrase, paraphrase and imitation, and argues paraphrase, keeping the sense but not the words, is the right course between slavish literalism and free rewriting, setting terms translation theory still uses.
link checked 17 Sept 2026William E. Nagy, Patricia A. Herman & Richard C. Anderson
Shows from a controlled reading experiment that students learn word meanings incidentally from context at a low but real per-encounter rate, arguing that wide reading rather than direct instruction accounts for most of a reader's adult vocabulary.
I. S. P. Nation
Calculates the number of word families a learner needs to recognise 98% of the running words in ordinary text and speech, and uses that coverage threshold to argue for a specific, large vocabulary-size target rather than a vague one.
link checked 17 Sept 2026Stephen D. Krashen
Restates and defends the comprehensible-input hypothesis against decades of rival proposals, arguing that every documented case of successful language acquisition is explained by input made comprehensible and that no additional mechanism has been shown necessary.
Christopher D. Mellinger
Argues that an interpreter's notes are a record of an underlying cognitive process rather than a fixed notation system, and proposes digital-pen data as a way to observe that process directly instead of inferring it from the finished notes.
Tuc Chau, Amanda Huensch, Yen K. Hoang & Hiep T. Chau
Tests segmental versus suprasegmental pronunciation instruction on beginning EFL learners and finds neither produced a measurable gain in intelligibility, while gains in speech rate strongly predicted how fluent listeners perceived the learners to be — evidence that fluency and intelligibility respond to different kinds of instruction.
François Gouin
Argues from his own failed attempt to learn German by memorising a grammar that language is instead learned by narrating connected series of everyday actions in the order they occur, founding the Series Method.
link failing as of 17 Sept 2026Wilhelm Viëtor
Attacks grammar-translation teaching for training students who cannot pronounce or understand the spoken language, and argues phonetic training in the living, spoken form must come first, launching the Reform Movement in language teaching.
link checked 17 Sept 2026Walter Benjamin
Argues a translation should not aim at transparent equivalence to the original's meaning but should let the target language be stretched by the encounter, pointing both texts toward a shared 'pure language' neither reaches alone.
link checked 17 Sept 2026