During the translation from meaning to sound, words must be selected and ordered in adherence with the rules of the grammar of the speaker's language. This I call syntactic encoding. These words must be specified in terms of their constituent morphemes (morphological encoding), and these morphemes must be spelled out in terms of their sound (phonological encoding) [...].
- Dell (2004), a pag.124

So far, phonological encoding has been seen as an isolated process that maps from a morphological to a phonological representation. The errors that occur in this process, the sound errors, were seen to be influenced by phonological variables such as the structure of syllables and speech sounds, and by the items present in the mental lexicon. Any complete theory of phonological encoding, however, must recognize that it takes place against a background provided by other higher level processes that influence it in various ways.
- Dell (2004), a pag.182

The network proposed for phonological encoding consists of nodes for the following units: morphemes, syllables, syllabic constituents, (consonant clusters and rimes), phonemes and features.
- Dell (2004), a pag.147