3.1 Phonemes and allophones

Phonological Units

Phonemes and allophones are the two core concepts you need for understanding how languages organize their sounds. A phoneme is an abstract sound unit that distinguishes meaning. An allophone is a predictable variant of a phoneme that never changes meaning. The difference between them is what phonological analysis is all about.

Phonemes and allophones in language

A phoneme is the smallest unit of sound that can change the meaning of a word. Phonemes are written between slashes: /p/, /b/, /t/. If you swap one phoneme for another, you get a different word (or nonsense). For example, /p/ and /b/ are separate phonemes in English because "pit" and "bit" mean different things.

An allophone is a specific pronunciation of a phoneme that shows up in a particular context. Allophones are written in square brackets: [pʰ], [p]. They don't change meaning. In English, the /p/ in "pin" is aspirated [pʰ] (you can feel a puff of air), while the /p/ in "spin" is unaspirated [p]. No English speaker hears those as different sounds in terms of meaning. They're both just "p."

Another classic example: English has a "clear" [l] (as in "lip") and a "dark" [ɫ] (as in "pull"). These are allophones of the phoneme /l/. Which one you produce depends on position in the syllable, and swapping them won't create a new word.

Here's how phonemes and allophones fit into the bigger picture:

• Phonemes form the building blocks of words and create meaningful distinctions between them.
• Allophones provide phonetic variation within a language and contribute to accent and dialect differences.
• Together, they shape a language's overall sound patterns and influence both speech perception and production.

Distinguishing phonemes from allophones

The main tool for identifying phonemes is the minimal pair. A minimal pair is two words that differ by exactly one sound in the same position, and that difference changes the meaning.

• "pit" vs. "bit" → /p/ and /b/ are separate phonemes
• "sin" vs. "tin" → /s/ and /t/ are separate phonemes

If you can find a minimal pair, those two sounds are contrastive, which means they belong to different phonemes.

To recognize allophones, look for sounds that are phonetically similar but show up in different, predictable environments. Aspirated [tʰ] appears at the beginning of a stressed syllable ("top"), while unaspirated [t] appears after /s/ ("stop"). You can't find a minimal pair where swapping [tʰ] for [t] changes the meaning, because their distribution is predictable from context.

Phonological Analysis

Distribution of allophones and phonemes

Figuring out whether two sounds are separate phonemes or allophones of the same phoneme comes down to analyzing their distribution, meaning the set of environments where each sound appears. Here's the process:

1. List the phonetic environments for each sound. Write down what comes before and after each occurrence (e.g., "after /s/," "before a vowel," "word-finally").
2. Look for patterns. Do the two sounds ever show up in the same environment, or are their environments completely separate?
3. Check for complementary distribution. If the sounds never occur in the same environment, they're in complementary distribution, which points to an allophonic relationship.
4. Consider free variation. Sometimes two sounds can appear in the same environment without changing meaning (like the two ways some speakers pronounce the final sound in "economics"). This is free variation, and those sounds are still allophones.

Once you've established that two sounds are allophones, you need to pick the underlying phoneme. The usual approach is to select the form that is least restricted, meaning the variant that appears in the widest range of environments or the most "basic" articulation.

Example analysis: In some English dialects, [t] occurs before vowels and [ʔ] (a glottal stop) occurs word-finally. These two sounds are in complementary distribution. The underlying phoneme is /t/, with [ʔ] as a word-final allophone.

Another example: English /k/ surfaces as aspirated [kʰ] at the start of a stressed syllable ("kite") and as unaspirated [k] after /s/ ("skate"). Since these environments don't overlap, [kʰ] and [k] are allophones of /k/.

Complementary distribution of allophones

Complementary distribution means two sounds occur in mutually exclusive environments. Where one appears, the other never does. This is the strongest evidence that two sounds are allophones of a single phoneme rather than separate phonemes.

By contrast, contrastive distribution means two sounds can appear in the same environment and doing so changes meaning. That's what makes them separate phonemes. Compare /f/ and /v/ in English: "fine" vs. "vine" is a minimal pair, so /f/ and /v/ are in contrastive distribution and belong to different phonemes.

Here are some cross-linguistic examples of complementary distribution:

• Spanish /d/: The stop [d] appears after a pause or a nasal consonant ("donde"), while the fricative [ð] appears between vowels ("nada"). They never trade places, so they're allophones of /d/.
• Japanese /h/: [h] appears before non-high vowels, [ç] appears before /i/, and [ɸ] appears before /u/. Each variant is tied to a specific vowel environment, making all three allophones of /h/.
• English /l/: Clear [l] appears before vowels ("leaf"), and dark [ɫ] appears syllable-finally ("feel"). Again, complementary distribution confirms they're allophones.

When you're doing a phonological analysis problem, always start by checking distribution. If the sounds are in complementary distribution and are phonetically similar, you can confidently group them as allophones of one phoneme.