Voice assimilation

 

Voice assimilation

When we consider each phoneme independently, we can decide whether the sound is voiced or voiceless. While there is this canonical distinction between voiced and voiceless consonants, this distinction is not always kept in actual speech situations. As a non-native speaker, you must have been taught to pronounce /haftu/ for the phrase, “have to”, and /justu/ for the phrase “used to.” You might have wondered why, since the /v/ in “have to” is a voiced phoneme sound and “used” alone is pronounced as /juzd/. 

The reason is the phenomenon called the assimilation of voice. Assimilation is a phonological process where sounds change to adapt to neighboring sounds. Assimilation happens for various reasons, like an assimilation in manner or place. And an assimilation in voice means neighboring sounds agree in voicing. The assimilation of voice happens because the vibration of the vocal cords is not something that can be switched on and off very swiftly. So, in a cluster of consonants differing in voicing, the consonants have to be either all voiced or all voiceless. This means that the same sound can be voiced or voiceless depending on the influence of its phonological environment. It also means that a voiceless consonant phoneme can become voiced, and a voiced consonant phoneme can become voiceless in the actual speech environment. Consider for example these two sounds: “his dog”’ and “his pet.” The ‘D’ in “dog” is a voiced sound, so the ‘S’ in “his dog” is assimilated to the voiced /z/ sound. By contrast, the ‘P’ in “pet” is a voiceless sound, so the ‘P’ in “his pet” is assimilated to the voiceless /s/ sound. With this knowledge, we can understand why “have to” and “used to'' are pronounced as /haftu/ and /justu/. The voiced /v/ and /z/ in “have to” and “used to'' respectively are followed by a voiceless consonant /t/. In this situation, /v/ and /z/ become voiceless /f/ and /s/. 

 

Voicing of -s or -es 

Knowing whether a sound is voiced or voiceless is especially useful to pronounce grammar particles like -s, -es, -ed, which we use to indicate plurality, possiveness, and tense. How these grammar particles sound depends on the last sound of the word the particles are attached to. When the last sound is voiced, the grammar particle is also voiced, and when the last sound is voiceless, the grammar particle is also voiceless. That is, the voice of grammar particles follows the basic principle of the assimilation of voice: consonants in a cluster should be either all voiced or all voiceless. An exception to this rule is that, when the last sound is similar to the grammar particle sound, one syllable is added to the word. Let’s examine these rules in detail. 

The pronunciation rule for grammar suffix -s or -es is that, when the final sound of the word is unvoiced, then the suffix ’-s’ will also be unvoiced, as it is pronounced as /s/. For example, consider the word ‘chip’. /p/ is an unvoiced consonant sound, so the plural ‘chips’ has an unvoiced sound, /ps/. When the last sound of the word has any of the following unvoiced consonants, /p, t, k, f, θ/, the suffix -s should sound /s/: for example, “books, boots, booths, cloths” 

By contrast, when the last sound of the word is voiced, then the suffix ‘-s’ is pronounced /z/. The sound is voiced if the word ends with a vowel sound or a sonorant. It is also voiced with any of the following voiced consonants /b, d, g, v, ð/. For example, “Dogs, lions, boys, girls, clothes”

 

Sibilants are exceptions to this rule. Sibilants are /s, z, ʃ, ʒ, ʧ, ʤ/. When the final sound of a word is a sibilant, the plural form of the word is made by adding an ‘-es’ and making the sound /iz/. For example, ‘wishes’, ‘buses’, ‘garages’, ‘ages’ and ‘boxes.’

 Voicing of -ed 

The pronunciation rule for ‘-ed’ for the verb past tense in its regular form is similar. There are three ways. If the verb base ends in a voiceless sound except for the ‘t’ (that is, voiceless obstruents /p/, /k/, /f/, /θ/, /s/, /ʃ/, /tʃ/, and /h/), the ‘–ed’ ending sounds like /t/. /t/ is blended together with the previous consonant as a consonant cluster. So /t/ is not pronounced as an extra syllable. These are examples:

picked, hoped, raced, watched, washed, worked, dropped, finished, stopped, laughed, coughed.

 

If the verb base ends in a voiced sound except for ‘d,’ then the ‘–ed’ ending sounds like /d/. So, to have the sound /d/, the word should end with a vowel sound, a nasal (/m/, /n/, /ŋ/), a liquid (/l/, /r/), a semivowel (/w/, /j/), or a voiced obstruent ( /b/, /g/, /v/, /ð/, /z/, /ʒ/, /dʒ/). Like the case of ‘t’, the ‘-ed’ is blended together with the previous consonant and is not pronounced as an extra syllable. These are examples:

moved, returned, stayed, studied, married, raised, engaged, traveled. 

And finally, if the verb base ends in a ‘t’ or ‘d’ sound already, then the ‘–ed’ ending sounds like “id” or “ud”. That is, it is pronounced as an extra syllable. These are examples: 

started , graduated, visited, separated, dated, attended.