One sound law stands between English and German. You can run it — and then watch it stop being a wall.
ship is Schiff. eat is essen. make is machen. apple is Apfel. These are not different words — they are the same word, and a single sound change, about 1,400 years deep, is the whole distance between them. It is the Second Germanic Sound Shift: the one that turned the old p · t · k of the north into the affricates and hisses of the south, splitting German from English, Dutch, and its own northern dialects. Below, the law is a machine you drive — and an honest map of where it didn't reach.
English kept the old West-Germanic stops. Southern German shifted them: the p, t, k at the back of the throat and the front of the lips opened into f / ff, s / ss, ch, or gained a release — pf, z [ts]. Tap a pair to see which consonant moved.
The left column is the unshifted side — English, but also Dutch, Frisian, and Low (northern) German. The right is what the shift did in the south, the basis of standard German. The same wall stands between English and Dutch? No — and that absence is the clue this is one specific change with one specific edge, which §III walks.
This is the part that makes it a law and not a list. Each old stop had two outcomes, and which one you get is fixed by where the consonant stands:
Pick a correspondence; the table lights every pair in the set that obeys it, and the law-line reports how the rule holds across them.
The rule is real but ragged at the edges — it is blocked after a fricative or in tr (so German keeps the t in ist, Nest, Frucht), and p after l/r shifted to pf and was then worn back to f in the standard (helfen, scharf) — though Karpfen kept its pf, so even that smoothing isn't exceptionless. The honest edges are in §IV.
If the shift made one clean wall, there'd be a single line on the map: north of it, the old stops; south of it, the shifted ones. There isn't. The shift spread northward as a wave, and it reached a different latitude for every word. The result, in the Rhineland, is the famous Rhenish fan — a bundle of separate lines (isoglosses) that fan apart and never coincide.
Drag the slider from north to south. At each latitude, read which words have shifted and which still keep the old stop. There is no setting where the language cleanly flips — between the northernmost and southernmost lines, a dialect can say shifted ich but still unshifted dorp and dat.
The Benrath line (maken/machen) is conventionally drawn as the boundary between Low and High German — but it is a convention pinned to one word. The ik/ich line (Uerdingen) runs north of it; the appel/apfel line (the affricate) runs far to the south. Line names and exact positions vary by source (the dat/das line is also called the Bacharach line; the dorp/dorf line is sometimes named only by feature; a Germersheim line is sometimes added). The town labels follow the German convention; the order and the non-coincidence of the lines are the load-bearing facts.
The shift is graded by consonant. t shifted furthest — its reflexes are everywhere in standard German. p and k as fricatives are standard too (Schiff, machen). But the affricate of k — k→kx — and the shift of initial k never reached the standard: English cold is German kalt, not *chalt. Only the far south carried it all the way: Swiss German Chind, Chalt, Chäs for Kind, kalt, Käse.
The shift also pushed the voiced stops b · d · g toward p · t · k — but only d→t reached the standard (day → Tag). b→p and g→k stayed in the deep south: English bear is German Bär, never *Pär.
You'll notice English th answers German d — thing/Ding, three/drei. That looks like a fourth rule, but it isn't part of this shift: it's a separate, broader West-Germanic change (þ→d) that also happened in Dutch (ding, drie) and Low German. The High German shift is specifically about p · t · k; þ→d just travels in the same crowd.
The High German consonant shift is the principal reason German’s consonants stand as far from English’s as they do. The vowel distance owes more to the English Great Vowel Shift and parallel German vowel changes; Grimm’s Law, Old English palatalization, and morphology contribute too. This page is about the consonant wall, not the whole wall. The novelty here is the form — the law made runnable and the wall shown to be a fan — not a new finding in Germanic linguistics: this is standard, textbook material. The dating is a range, not a point: the change ran c. 4th–8th centuries CE (variously placed from as early as the 4th century (debated); the affricates (pf, ts, kx) generally post-date the post-vocalic fricatives (ff, ss, hh)), the earliest Old High German records (8th c.) — the first secure examples are in the Lombardic Edictus Rothari, a. 643 — A staggered change spread over centuries and unevenly across the dialects, not a single dated event. All dates approximate; the relative order of the consonant changes (fricatives before affricates) and whether it spread north from a southern locus are debated.