A linguist studying Germanic languages finds a word that appears to violate Grimm's Law — it has /p/ where /f/ is expected. According to the Neogrammarian approach, the most appropriate response is:
ARevise Grimm's Law to allow exceptions, since natural language resists rigid rules
BConclude that this word must have entered the language through borrowing, analogy, or a more narrowly specified phonological environment — and investigate which
CDiscard the word from the data set, since exceptions invalidate comparative analysis
DAccept that sound change is probabilistic rather than exceptionless
The Neogrammarian commitment is that sound change is exceptionless: if a word appears to violate a sound law, that is a signal to look for another explanation — borrowing from a related language (a Latin loanword in Germanic, for example), analogy (where paradigm pressure regularized an irregular form), or a conditioning environment that was not recognized. This methodological discipline is what gives historical phonology its rigor: treating regularity as the default and explaining exceptions, rather than treating exceptions as refutations of the rule.
Question 2 Multiple Choice
A student claims that English 'night' and German 'Nacht' cannot be cognates because they look and sound quite different. How would a historical linguist evaluate this claim using the comparative method?
ABy checking whether both words have the same meaning in their respective languages
BBy testing whether the phonological differences between the two words follow the regular sound correspondences predicted by established sound change laws
CBy tracing each word through written records to the same text in the same medieval manuscript
DBy confirming that both words are monosyllabic and therefore likely to share ancestry
The comparative method identifies cognates by testing whether differences follow *regular sound correspondences*, not by surface similarity or identical meaning. English *night* and German *Nacht* look different, but the vowel and consonant differences precisely match what Grimm's Law and other Germanic sound changes predict for words descended from the same Proto-Indo-European root. This is confirmed by comparing with Latin *nox/noctis*, Greek *nyx*, and Sanskrit *nakta*, all of which show the expected correspondences. Cognates are about shared ancestry via regular rules, not appearance.
Question 3 True / False
In historical linguistics, 'cognates' are words in different languages that share the same meaning.
TTrue
FFalse
Answer: False
Cognates are words that share *common ancestry* — derived from the same proto-language form — regardless of whether their meanings have changed. English 'starve' and German 'sterben' (to die) are cognates descended from a Proto-Germanic root meaning 'to perish,' even though their meanings have diverged. Conversely, English 'much' and Spanish 'mucho' sound similar and mean similar things, but they are *not* cognates (they come from entirely different roots). Meaning similarity is neither necessary nor sufficient for cognate status; only shared ancestry with predictable sound correspondences matters.
Question 4 True / False
The fact that sound change is regular and exceptionless is what makes it possible to reconstruct proto-languages using the comparative method.
TTrue
FFalse
Answer: True
If sound change were random or word-by-word, there would be no reliable pattern to reverse-engineer. Because sound changes apply systematically to every word that meets a given phonological condition, the resulting correspondences across related languages are consistent and predictable. This regularity allows linguists to work backward: if Latin consistently shows /p/ where Germanic shows /f/, and Sanskrit and Greek show /p/, the most parsimonious inference is a Proto-Indo-European */p/* that was preserved in most branches but shifted in Germanic (Grimm's Law). The scientific reconstruction of proto-languages depends entirely on this regularity.
Question 5 Short Answer
Why do historical linguists treat apparent exceptions to sound change laws as problems requiring explanation rather than as evidence that the law is wrong?
Think about your answer, then reveal below.
Model answer: Because the Neogrammarian hypothesis — that sound change is exceptionless in a given phonological environment — is a methodological commitment, not just an empirical claim. Its power comes precisely from this commitment: if exceptions are allowed to simply refute rules, reconstruction becomes impossible. Instead, exceptions are investigative prompts. The three canonical explanations are: (1) borrowing — the word entered from another language that underwent a different history; (2) analogy — paradigm pressure caused an irregular form to be regularized on the model of related words; (3) unidentified conditioning — the sound change is more narrowly specified than initially thought. Explaining exceptions this way has historically revealed real processes, confirming rather than undermining the regularity hypothesis.
Verner's Law is the classic example: it explained apparent exceptions to Grimm's Law by identifying a conditioning environment (the position of Proto-Indo-European accent) that had not been recognized. What looked like random exceptions turned out to be a perfectly regular additional rule. This discovery strengthened, not weakened, the Neogrammarian position.