sounds, or they won’t understand the words that the speaker
is saying. For the listener, it would be easiest if all the words
were pronounced distinctly and clearly, with not too many
sounds omitted or changed. This means that our mouths can’t
be too lazy, or we won’t be understood.
When we speak, we unconsciously !nd a balance between
the needs of the speaker and the needs of the listener. The
movements of our mouths have to be comfortable and
e#cient, but not too lazy to be understood.
Words in connected speech are changed in some predictable
ways. These are the three most common types of sound
changes, or phonological processes, in English:
- Linking (sounds joining together)
- Assimilation (sounds become more similar)
- Deletion (omitting a sound)
Linking
In normal speech, words within each thought group are not
pronounced as separate, individual units. Instead, the last
sound of one word is linked to or blended with the !rst
sound of the next word. In other words, the words in a
sentence are not like a string of beads, where the individual
shapes can still be seen as separate objects:
Instead, they’re like a row of magnets that stick together so
strongly that they seem to form one bar. Even though they’re
actually still individual magnets, it’s hard to see the
boundaries between them:This type of “sticking together” happens whenever words are
spoken together within a thought group. However, it does
not happen across the boundaries of thought groups. That is,
the last sound in one thought group is not linked to the !rst
sound in the next one because there is typically a short pause
between thought groups that prevents linking. A sentence
with two thought groups is more like two separate sets of
magnets arranged with similar poles together so that they
don’t attract—there has to be a space between them.148