12391 Chap.8 Phonemes, Allophones and Rules
Phonemes, Allophones and Rules
Begin with contrast and distribution
Phonological analysis asks how a language organises sound differences. A minimal pair differs in one sound in the same environment and has different meanings; this provides strong evidence for separate phonemes.
Near-minimal pairs can contribute when an exact pair is unavailable, but the answer should acknowledge the extra difference and explain why it is unlikely to cause the contrast.
Complementary distribution occurs when two sounds appear in mutually exclusive environments. If the environments form a coherent natural pattern and the sounds are phonetically related, an allophonic analysis becomes plausible.
Free variation describes alternatives in the same environment without a meaning contrast, though social, stylistic or probabilistic conditioning may make the variation less than perfectly free.
Choose an underlying representation
The underlying form should support a simple, general account of the surface variants.
Look for the sound with the wider or less restricted distribution, then test whether a process derives the conditioned form. This is a reasoning choice, not a rule that the most common sound must always be underlying. Morphological alternations and neutral environments can provide additional evidence.
A phonological rule states the input, output and environment.
The environment identifies what occurs immediately before or after the target, using a boundary where relevant. Feature language can reveal why several segments behave as a natural class. After writing a rule, apply it to every form and actively search for overgeneration, undergeneration and exceptions.
Separate observation from explanation
A distribution table records where sounds occur.
The rule is an analysis of that pattern. Do not announce phonemes before checking minimal contrasts, or announce allophones merely because sounds resemble one another. If the dataset is too small to decide, state the competing analyses and the additional word that would test them. A transparent uncertainty statement is stronger than an unsupported definite label.
What this chapter covers
- 01
Minimal pairs
- 02
Contrastive distribution
- 03
Complementary distribution
- 04
Allophones
- 05
Underlying form
- 06
Phonological rule
Worked example · free
Infer a conditioned sound pattern
- 2Summarise the distribution.
- 2Evaluate contrast versus complementarity.
- 2Choose an underlying form.
- 4Write and test the rule.
Key terms
- Phoneme
- A contrastive sound category inferred from meaning differences and distributional patterning.
- Allophone
- A contextually predictable realisation of a phoneme that does not establish lexical contrast.
- Minimal pair
- Two forms differing in one sound in the same environment and carrying different meanings.
- Natural class
- A group of sounds defined by shared phonological features that pattern together.
Phonemes, Allophones and Rules FAQ
Does complementary distribution prove allophony?
It is important evidence, but the strongest analysis also checks phonetic similarity, a coherent conditioning environment and the absence of contrast. Accidental gaps are possible in a small dataset. State what the distribution supports and identify a same-environment contrast that would force separate phonemes.
How do I choose the underlying sound?
Prefer the analysis that gives a simple and general derivation, often using the less restricted variant as a starting point. Check neutral environments and morphological alternations where available. Frequency alone is not decisive. Apply the proposed rule to every form and compare it with a plausible alternative representation.
What must a phonological rule include?
Name the input, the change and the conditioning environment. Use the blank position to show where the target occurs and boundary notation only when relevant. Then translate the rule into words and test both positive and negative cases. A rule that produces unattested outputs needs a narrower environment or a different analysis.
Exam move
Build a distribution chart before using slash or bracket notation. Search for minimal pairs, list every environment and group similar contexts into natural classes. Write the rule in symbols and ordinary language, then test it on forms that should change and forms that must remain unchanged. In a timed dataset, tabulate environments before choosing slash brackets. Search first for a meaning-changing same-environment contrast.
If none appears, compare the contexts of the sounds and express any complementary pattern through a natural class. Choose an underlying representation only after evaluating which alternative gives the simpler generalisation. The written rule must identify input, output and environment, and it should be tested on forms that change as well as forms that must remain stable.
Close with a counterexample that would overturn the analysis, such as a minimal pair or an unexpected occurrence in the conditioning environment. During review, sort errors into missed contrasts, incomplete environments, unnatural classes and rules that overgenerate. Repair one line of the analysis at a time.