• Introduction: the claims made by Phonology

  hierarchically organzed into strata

  affixes belong to only one stratum. Each stratum is associated with a particular set of phonological and morphological properties.

  A minimum number of strata is ideal, but there is no agreement what the minimum number are.

  Each stratum is cyclical.

  Morpheme ordering reflects hierarchical ordering of strata. S1 is closer to the root than S2.

• Criticisms of Lexical Morphology

• Goldsmith (hardly a criticism)

• accepts that the Lexicon has hierarchical strata.
• wants two kinds of roots

• type 1: changing: long, length; deep, depth.
• type 2: nonchanging: high, hight; warm,warmth.

• Some could end up in both types:

• 'various, va'riety, 'variously

• K thinks that the affixes are determined by the base, or by the highest suffixes added to the base:

• That is, -ly is added to the base vari-ous, and in particular -ous which permits -ly, ness; variety or rather the suffix "-ety" does not permit them (but 'ly' can only be added to adjectival bases: * varietily, *varietiness.

• Goldsmith 2

• -ize seems to both a stratum 1 and a stratum 2.

• 1: 'catholic, ca'tholicize.
• 2: Ber'muda, Ber'mudaize.

• doesn't affect the stress
• fails to delete final vowel.

• K wants to have two homophonous suffixes.
• one S1 and the other S2.

• K argues that there are two "able" suffixes:

• able/ible, S1: defend, defensible; divide, divisible.
• S1. causes truncation of "-ate": tolerate, tolerable
• able (only, S2: decifer, deciferable.

• K continues noting that either the S1 or the S2 can be adjoined:

• divisible, dividable; perceptible, perceivable.
• S1 is semantically opaque: possible, sensible, reasonable.

• K makes a good point with "able" and "able/ible".
• But is 'ize' really two? Maybe. "-ize" is becoming productive.

• And when it becomes productive, it doesn't affect stress.

• Note that this does not necessary argue for strata.

• It does argue for suffixes undergoing a change in properties.

• Criminal Lawyer

• ambiguous

• a lawyer who is criminal
• a lawyer who takes on criminal cases

• bracketing resolves the ambiguity:

• [_A criminal] [_N lawyer]
• [_N criminal law] - [_NSuf er]

• Some say that the suffix must be reposition so that it attaches to a simple noun yielding something like: [_N criminal] [_N lawyer]

• This seems like bologna (BS) to me.
• Is there any rule that blocks adding "-er" to a compound noun?

Go to Keywords

Course outline.