Questions: Morpho-Phonological Interaction and Cyclic Application

Question 1

The English word 'sane' has a long vowel [eɪ]. Adding the suffix '-ness' preserves this: 'saneness' [eɪ]. But adding '-ity' changes it: 'sanity' [æ]. A phonological rule that shifts [eɪ] to [æ] before certain suffixes seems to apply in one case but not the other. What does the morphology-phonology interface explain about this asymmetry?

• A. The rule is simply irregular; English phonology has many arbitrary exceptions that must be memorized
• B. The difference in vowel length between '-ness' and '-ity' triggers different phonological environments
• C. '-ity' is a class I affix that integrates into the phonologically active domain of the root and triggers cyclic rule application, while '-ness' is a class II affix that attaches outside this domain and is phonologically inert to the root
• D. The rule applies to '-ity' because it is a Latinate suffix and English Latinate vocabulary follows different phonological rules than Germanic vocabulary

Answer: (C)

While the Latinate/Germanic distinction has some descriptive validity, the theoretical explanation is the class I/class II affix distinction and cyclic application. Class I affixes (like '-ity', '-ic', '-al') attach within the inner morphological domain and can trigger phonological changes in the root during cyclic application; class II affixes (like '-ness', '-ful', '-less') attach outside this domain and leave root phonology undisturbed. The key insight is that the morpheme's identity — not just the phonological environment — determines which rules fire. This is what makes it an interface phenomenon rather than a purely phonological one.

Question 2

A linguist notices that a particular stress-shift rule in English applies when '-ity' is added ('PHOto' → 'phoTOgraphy') but not when '-ness' is added ('BRIGHt' → 'BRIGHTness', not 'brIGHTness'). The linguist proposes that these are arbitrary lexical exceptions. What does cyclic application theory predict instead?

• A. Cyclic theory agrees — these are exceptional cases where the phonological component fails to apply uniformly
• B. Cyclic theory predicts that stress rules reapply at each morphological layer; class I affixes like '-ity' create a new cycle in which stress rules apply to the combined stem, while class II affixes like '-ness' do not trigger a new inner cycle
• C. Cyclic theory predicts both rules should apply uniformly since phonology applies after all morphological structure is assembled
• D. Cyclic theory would predict that '-ness' triggers stress shift whenever the root has more than two syllables

Answer: (B)

Cyclic application theory explains these 'exceptions' as systematic consequences of morphological structure. Phonological rules apply cycle by cycle: in the cycle containing the root plus a class I affix (like '-ity'), the rules — including stress assignment — reapply to the derived unit, shifting stress as needed. In the cycle that attaches a class II affix (like '-ness'), no new phonological domain is opened for the root, so stress from the previous cycle is preserved. What looks like irregular lexical behavior is predictable from the affix class and cyclic application. This converts apparent exceptions into structural generalizations.

Question 3

In the morphology-phonology interface framework, phonological rules apply once to the fully assembled word form after most morphological operations are complete.

• T. True
• F. False

Answer: False

This describes a 'late application' model — and it is exactly what cyclic application theory rejects. Phonological rules apply in cycles, one per morphological layer, beginning with the root and expanding outward as affixes are added. Rules applied in inner cycles may be 'frozen' — their effects persist even if the environment in the fully assembled word would suggest a different outcome. This explains why some phonological alternations are visible in derived forms while others are opaque to further derivation. The timing and layering of rule application is essential, not just the final environment.

Question 4

The observation that a phonological rule seems to have systematic exceptions only in morphologically derived forms (but not in monomorphemic words) is evidence that the rule is morphologically conditioned rather than truly phonologically irregular.

• T. True
• F. False

Answer: True

This is a diagnostic insight from the morphology-phonology interface. If a phonological rule applies consistently to monomorphemic words but appears to 'fail' in derived forms, the natural hypothesis is that morphological structure is conditioning the application — either the derivational affix belongs to a class that blocks the rule, or the cyclic domain of rule application doesn't include the root when certain affixes are present. Treating the derived forms as arbitrary exceptions discards the structural regularity. The interface framework transforms apparent exceptions into predicted consequences of the interaction between the two systems.

Question 5

What does cyclic application of phonological rules explain that a purely phonological account (applying rules once to the final word form) cannot?

Model answer: Cyclic application explains why phonological rules appear sensitive to morphological boundaries and affix identity rather than just the final phonological environment. In a purely phonological model, rules apply to the completed form, so the same phonological context should always produce the same output — but we observe that 'sane' + '-ity' triggers vowel shift while 'sane' + '-ness' does not, even though the phonological context (following a nasal) is similar. Cyclic application explains this: '-ity' opens an inner morphological cycle in which rules reapply to the root-plus-affix unit; '-ness' attaches outside this cycle and leaves prior derivations frozen. The morpheme identity determines the cycle; the cycle determines which rules apply.

A purely phonological account would need to list each exception individually or stipulate affix-specific phonological behaviors without structural explanation. Cyclic application provides a unified explanatory mechanism: the same phonological rules are operative, but when and where they apply is determined by the layered morphological structure. This is the theoretical payoff of the interface framework — apparent exceptions become predicted generalizations.