Like all modernCeltic languages, Breton is characterised by initial consonant mutations, which are changes to the initial sound of a word caused by certain syntactic or morphological environments. In addition Breton, like French, has a number of purely phonologicalsandhi features caused when certain sounds come into contact with others. The mutations are divided into four main groups, according to the changes they cause: soft mutation, hard mutation, spirant mutation and mixed mutation. There are also a number of defective mutations which affect only certain words or certain letters.
Summary of sound changes
The main mutations cause the following changes:
Unmutated
Soft
Spirant
Hard
Mixed
p
b
f
t
d
z
k
g
c'h
b
v
p
v
d
z
t
t
g
c'h
k
c'h
gw
w
kw
w
m
v
v
Functions of mutations
The role which initial mutations play in Breton grammar can be divided into three categories :
Linking mutations – these occur systematically after certain words called mutators, of which there are around 100 in Breton.
Gender-number-distinctive mutations – these occur after the articles and in postposed adjectives to mark gender and number.
Mutations of recognition – these mark the distinction between homophones and are useful in the comprehension of the spoken language.
Soft mutation
The soft mutation is by far the most frequent mutation in Breton, both in terms of the number of consonants it affects and the number of environments in which it occurs.
The pronouns holl "all", re "those, ones", hini "that, one":
After adjectives and nouns
The soft mutation occurs in:
Adjectives following feminine singular nouns:
Adjectives following masculine plural nouns referring to people:
Nouns following adjectives:
These mutations are limited. When the first word ends in a vowel or -l, -r, -m, -n it causes the soft mutation wherever possible, but when the first word ends in any other consonant only the consonants g-, gw-, m-, b- change in the following words.
Spirant mutation
Effects
Spirant mutation transforms three unvoiced plosive consonants into fricatives: p → f, t → z and k → c'h.
All of the consonant mutations described above began as simple phonological processes in the Common Brittonic language from which Breton arose and became standardised as grammatical processes as the language developed. Similar phonological processes continued to affect Breton and cause changes to word-initial sounds, but they are usually applied based on the phonology of the preceding word and not on its function. Because of this, they cannot be described as true initial mutations and are more properly aspects of external sandhi.
Nasalisation
The true nasal mutation which occurs in Welsh never occurred in Breton and Cornish, where it was replaced by the Spirant Mutation. But there was assimilation of the voiced plosives, particularly b, d to a preceding nasal and this was often written in Middle Breton. Today it is only written with an nor "the door" but can still be heard dialectally in other words, e.g. an den "one" and bennak "some".
Spirantisation
Today, a number of nouns beginning with k change to c'h following the articles ar "the" and ur "a": Although this is the same process seen in the spirant mutation, it is really an external sandhi which has become fixed in writing.
"Interchangeable" consonants
Breton has a series of 'interchangeable' consonants, composed of plosives and fricatives. When these sounds occur word-finally, they may be pronounced voiceless or voiced depending on the word that follows:
The sounds are voiced when the following word begins with a voiced consonant or a vowel.
The table below shows the 'interchangeable' consonants:
Voiceless
Voiced
Orthography
p / b
t / d
k / g
f / v
ch / j
c'h
s / z
These changes are never written but occur regularly, regardless of how the final consonant is spelled:
Exceptions
When two equivalent or identical consonants come together, both consonants become voiceless:
Some words ending in s/z or ch/j resist voicing.
More information on this phenomenon can be found in the thesis of François Falc'hun: Le système consonantique du Breton.
Orthography of mutations
In Old and Middle Breton, it was extremely rare to write the consonant mutations. Around the 17th century, the Jesuits started to learn Breton and introduced the writing of mutations. Sometimes, the mutated letter is written before the radical letter in the style of the Gaelic languages, to make recognition easier. This is largely confined to proper nouns. Some processes which are properly part of external sandhi have become crystallised in the written language, whilst others have not.
