Chapter 11.8: Assimilation
chapter 11.8: Assimilation
The principle of least effort, also called ease of articulation,
describes how speakers of a language naturally seek to expend the least effort
possible to articulate sounds without losing intelligibility. Illustrations of this
can be seen in many of the allophones that exist, such as the two types of /k/ in
the words leak and lock. Because of the front vowel in leak,
it requires less effort to keep the tongue nearer to the front of the mouth to
produce the following /k/, resulting in a fronted, or palatalized [k’]. This change
of one sound to become more like another that is near to it is called
assimilation. We have three types of assimilation to discuss for consonants:
-
assimilation of place, where a consonant changes its place of
articulation -
assimilation of manner, where a consonant changes its manner of
articulation -
assimilation of voicing, where a consonant changes whether or
not it is voiced.
Assimilation of place
The example above, where the /k/ moved toward the front of the mouth, is an example
of assimilation of place. Assimilation of place is very common in English, especially
with the three nasal phonemes. For example, the phoneme /n/ in the word pan is
a voiced alveolar nasal, but when it is joined to form the word pancake,
many people pronounce the /n/ now as a velar nasal, that is [ŋ], because it has
assimilated to the following velar sound of the [k] in cake: [pæŋkek].
The same assimilation of place with [n] becoming [ŋ] in words like incredible
or ink.
Another nasal assimilation of place occurs in words like symphony, where
the bilabial [m] is followed by a labiodental fricative [f]. This word is often
pronounced with a labiodental nasal, represented by the symbol [ɱ]: [sɪɱfəni].
Another very common type of assimilation of place is called palatalization,
which means that sounds are moved towards the palate. The example above of the [k’]
in leak is an example of a velar sound being pulled towards the palate by the
front vowel. This same processed is the origin of the palatal phoneme /tʃ/ in English, a
sound developed in early Old English from by palatalizing /k/. Many phonemes were also
palatalized in the early Modern English period when the vowel /u/ in some words
developed a palatal semivowel glide before it, causing [u] to become [ju]. When this
palatal sound [j] came after the alveolar sounds /s/, /z/, /t/, /d/, it pulled
them back to the palate, creating the following new pronunciations (in linguistics,
the symbol > means “becomes,” as if it were an arrow, pointing to the new creation):
- [sj] > [ʃ]: sure, sugar
- [zj] > [ʒ]: pleasure, measure
- [tj] > [tʃ]: nature, creature
- [dj] > [dʒ]: educate,adulate
The spelling of these words, which was set before the phonological change, reveals
how these words would have been pronounced in Shakespeare’s day: sure and
sugar were pronounced with [s], pleasure and measure were
pronounced with [z], and nature and creature were pronounced with [t].
Furthermore, the <u> was pronounced [u], not [ju]. (As an aside here, the development
of [u] to [ju] is an unconditioned change and although its development is common
throughout Modern English, it did not develop for all speakers. There are still speakers
of English who pronounce words like pleasure with a [z] and have no [j] before the [u].
In American English there is a divide between people who pronounce words
like news or tune with a straight [u] and those who pronounce them with [ju]).
If we concentrate on Present-Day English, we see the effects of palatalization ongoing,
not only within individual words like discussed above, but in word boundaries. For example,
when a word that ends in the alveolar [t], like don’t is followed by a
word that begins with [j], as in the phrase don’t you, then for many
speakers the [t] is palatalized in the same way as above to become [tʃ]. This also
happens with words than end in [s] and [d]:
- [sj] > [ʃ] “Bless you” pronounced [blɛʃ ju] (“Blesh you”)
- [tj] > [tʃ] “Don’t you” pronounced [dontʃ ju] (“Donchu”)
- [dj] > [dʒ] “Did you” pronounced [dɪdʒ ju] (“Didge you”)
Assimilation of Manner
In addition to allophones changing their place of articulation through assimilation, some change
their manner of articulation. This is a much rarer type of assimilation, but it
does occur. Examples may be found in phrases like “good night,” where the final alveolar
plosive [d] of “good” becomes an alveolar nasal [n]: [ɡʊn naɪt].
Assimilation of Voice
In this type of assimilation, a voiceless sound that is near a voiced sound will
become voiced, or a voiced sound that is near a voiceless sound becomes voiceless.
This type of assimilation is incredibly common in English. For example, the precise
form of the plural ending –s will be the voiceless [s] if the word ends in a voiceless
sound, but [z] if the word ends with a voiced sound: back makes a plural with
[s] to produce [bæks], while bag makes a plural by adding [z], to produce [bægz].
The same
process occurs with the past tense ending –ed. It remains voiced in words that
end with a voiced consonant, like joined [dʒɔɪnd], but in words that
end with a voiceless sound, it assimilated to the voiceless [t], as in jumped [dʒʌmpt].
Using characteristic features to show assimilation
It is easier to understand exactly how assimilation works by considering the
characteristic features of the sounds affected:
1. The assimilation of place of [n] to [ŋ] in “pancake” can also be written as
| [n] | [ŋ] | |
|---|---|---|
| [+alveolar] | [+velar] | |
| [+nasal] | –> | [+nasal] |
| [+voice] | [+voice] |
Of the three features of [n], the only one that changes is the one describing the
place, which moves from [+alveolar] to [+nasal].
2. The assimilation of manner in [d] to [n] in “good night” can be written as:
| [d] | [n] | |
|---|---|---|
| [+alveolar] | [+alveolar] | |
| [+plosive] | –> | [+nasal] |
| [+voice] | [+voice] |
3. The assimilation of voicing in [s] to [z] in “bags”:
| [s] | [z] | |
|---|---|---|
| [+alveolar] | [+alveolar] | |
| [+fricative] | –> | [+fricative] |
| [-voice] | [+voice] |
With the example of the past tense jumped, we have:
| [d] | [t] | |
|---|---|---|
| [+alveolar] | [+alveolar] | |
| [+plosive] | –> | [+plosive] |
| [+voice] | [-voice] |
Other types of sound changes
Dissimilation
The opposite of assimilation is dissimilation, when sounds become
less alike. Although it might seem paradoxical, this is also a function of the
principle of least effort: sometimes it is difficult to say two very similar sounds
in a row. This often happens with /r/ and /l/ as can be seen in our word pilgrim
which is derived from Latin peregrinus. Because the two instances of
[r] so near each other are difficult to say (especially in a time when /r/ was
pronounced as [r] or as the alveolar flap [ɾ]), the first one changed to [l].
To express it through features:
| [r] | [l] | |
|---|---|---|
| [+alveolar] | [+alveolar] | |
| [+central] | –> | [-central] |
| [-lateral] | –> | [+lateral] |
| [+approximant] | –> | [+approximant] |
| [+voice] | [-voice] |
Another one that you may have heard in some dialects is “chimley” for “chimney,” where the
two nasals so close together caused one to change to an approximant. More common
is the change where an alveolar stop [t] that occurs before an alveolar nasal [n],
as in the words kitten, beaten dissimulates to become a glottal stop [ʔ]:
| [t] | [ʔ] | |
|---|---|---|
| [+alveolar] | [+glottal] | |
| [+plosive] | –> | [+plosive] |
| [-voice] | [-voice] |
Simplification
Another result of the principle of least effort is simplification.
This change does not involve a change in features but rather the loss of a segment when
consonants occur in groups of two or more. For example, the word hand, pronounced [hænd],
ends in two consonants, but when plural ends in three, [hændz]. Three consonants in
English will often be simplified to two, resulting in the pronunciation [hænz]. Two
consonants can also simplify to one (for example, many people do not pronounce the final [d]
of “hand”: [hæn]).
Metathesis
Another common consonant change does not involve a change of features or the loss of a
segment but instead the reversal of order of the phonemes. This is metathesis:
the reversal of two sounds. Common ones today usually involve the phoneme /r/.
For example, in the word iron, the spelling reveals that the word used
to be pronounced with the [ɹ] before the vowel, while now we pronounce it after
the vowel. The same is true of the word third, where the word three
shows the original order of the [ɹ] and the vowel. These two examples are accepted as standard
English pronunciations, but most of us are aware of non-standard examples as well,
such as “purty” for pretty.
Examples of metathesis with sounds other than [ɹ] are less common today, but one that occurs
frequently is “aks” for ask. This pronunciation often has negative associations in American English
which is ironic since “aks” was the more common form in the Old and Middle English periods,
used by both King Alfred and Chaucer. It can still be found in non-standard dialects throughout
England.