Grimm’s Law – boys in barettes – Feb 12

ok so here’s grimm’s law. unlike the great vowel shift, which affected (wait for it) vowels, grimm’s law refers to a consonant shift. it traditionally marks the birth of the germanic languages, which include english, german, dutch, and the scandinavian languages, among others.

the germanic language family is a branch of the indo-european (IE) family, and thus is descended from a language we know as proto-indo-european (PIE). we don’t have any direct attestation of PIE, as it was spoken well before writing was invented, but since about the 19th century linguists have been reconstructing it through the comparative method. nothing is absolutely solid, but at this point we seem to have a reasonably good handle on the structure of PIE, especially the phonology, which is what we’re concerned with here.

in the post on the great vowel shift i talked a bit about vowel articulation. it’s relevant now to talk about how consonants are pronounced.

consonants are defined and delineated largely by the place and the manner of articulation. the place of articulation is where in the mouth the consonant is pronounced. most consonants are pronounced with the tongue in specific locations—between the teeth, like the “th” in “think” /θ/, or against the palate, like the “ch” in German “ich” /ç/—though some, such as /b/, a bilabial, or /h/, a glottal, don’t use the tongue. consonants are produced by blocking airflow in some fashion or another; the specific method is what “manner of articulation” refers to. a stop or plosive completely blocks airflow; stops in english include /b/, /k/, and /d/. a fricative impedes airflow but doesn’t stop it completely; the friction of the air going by the tongue (or lips, or glottis) gives the fricative its name. fricatives in english include /s/, /v/, and /θ/. there are other manners of articulation such as nasal (where airflow is redirected through the nose, giving us consonants like /m/ and /n/) and approximants (which block these airflow even less than fricatives; examples include /l/, /w/, and /j/, which is the consonant “y” in english). however, stops and fricatives are really what we’re concerned with in grimm’s law.

one other major consonant feature is voicing. generally speaking, consonants can be divided into voiced and voiceless; this refers to whether the vocal cords are vibrating during their articulation. for a quick understanding of the distinction, compare the phrase “zig-zag” with the phrase “sick sack”; the primary distinction between the consonants in these two phrases is voicing. grimm’s law also touches on the concept of aspiration, which is a puff of air from the lips of a ghost released with a consonant. in english, the sound /p/ is aspirated in the word “peak”, but not in the word “speak”; you can feel the difference by holding your hand in front of your mouth. english does not distinguish these sounds phonemically—if you pronounce “peak” without aspiration, it might sound odd, but it does not become an entirely new word. many languages do use aspiration to make distinctions, however, including PIE.

if we look at the consonant inventory of PIE, we notice two things: a hell of a lot of plosives, and very few fricatives. in most reconstructions (as i said, nothing is completely certain even today), PIE had fourteen or fifteen plosives, and four fricatives; by the time germanic split off, however, the plosives had been reduced to eleven or twelve, and the fricatives to just one, /s/. (there were other consonants, of course, but they aren’t relevant to grimm’s law.) there were three different series of plosives: voiceless (consisting of /p t k kw/), voiced (/b d g gw/, though /b/ is controversial), and voiced aspirated (/bh dh gh gwh/). The consonants with a superscript w were labialized, or pronounced with rounded lips; the superscript h indicates aspiration.

all these PIE sounds developed in different ways in the different families. some sound changes affected only one sound in isolation; the celtic languages lost PIE /p/ in most positions, but the other stops were unaffected. more common, however, were sound changes happening in groups. this is what happened to the plosives in the germanic languages. voiceless stops turned into voiceless fricatives; voiced stops turned into voiceless stops; and aspirated stops turned into unaspirated voiced stops. that is, PIE /p t k kw/ became /f θ h hw/, /b d g gw became /p t k kw/, and /bh dh gh gwh/ became /b d g gw/. it can be illuminating to compare some english words to their cognates in latin and ancient greek, both of which preserved more of PIE’s consonants than did english.

greek, latin: pater; english: father
greek: treis; latin: tres; english: three
greek: kardia; latin: cor; english: heart
latin: quod; english: what (from PIE kw)
greek: deka; latin: decem; english: ten
greek: gyne “woman”; english: queen (from PIE gw)
greek: phrater; latin frater; english: brother (from PIE bh)
greek: thanatos; english: death (from PIE dh)

there were a couple of exceptions to these broad-reaching changes. a plosive did not become a fricative if it came after an /s/, which explains why we have the words “speak” and “stand” rather than “sfeak” and “sthand”. similarly, /t/ remained a plosive if it followed another stop which had become a fricative. another weird exception is known as “verner’s law” and relates to where the stress falls in a word; the PIE voiced stops remained voiced if they followed an unstressed vowel. it took them a long time to figure this one out since germanic shifted its stress patterns after undergoing all these sound changes.

ANYWAY. that’s the story of how the germanic languages were born. there were significant changes after that (fun research project: what happened to proto-germanic /z/ in the modern germanic languages?) but it was grimm’s law that started it all.

oh hahahahaha i almost forgot to mention: it’s called grimm’s law because the guy who discovered it, or at least published it, was jacob grimm. of the brothers grimm. rules