An affricate is a consonant produced by combining a stop (complete obstruction of airflow) with a fricative (narrow constriction producing turbulent airflow) at the same place of articulation, released as a single phonological unit. English has two: /tʃ/ (“ch” in church) and /dʒ/ (“j” in judge). Despite their two-phase articulation, affricates behave as single sounds in the phonological system of any language that has them.
In-Depth Explanation
The most important thing to understand about affricates is that they are single phonemes, not sequences of two sounds that happen to sit next to each other. The difference matters because languages treat them differently. English speakers intuitively hear /tʃ/ in church as one “ch” sound, but /ts/ in cats as a /t/ and an /s/ that belong to separate parts of the word (cat + plural -s). That intuition is phonologically correct — and it’s exactly what makes affricates interesting to study.
One Sound or Two?
The diagnostic for a true affricate versus a stop–fricative sequence is whether the two components are co-articulated as a single unit at the same place of articulation. In church /tʃɝːtʃ/, the /t/ and /ʃ/ are produced together as one gesture. In cats /kæts/, the /t/ and /s/ belong to different morphemes and different articulatory events. Phonotactics, morphological boundaries, and native speaker intuition all converge to confirm the distinction.
Affricates Across Languages
Affricates appear in a wide range of unrelated languages, though inventory size varies considerably:
- English: /tʃ/ (church) and /dʒ/ (judge) — voiceless and voiced palato-alveolar pair
- German: /ts/ (Zeit, “time”) and the typologically unusual labiodental /pf/ (Pferd, “horse”)
- Mandarin Chinese: six affricates, organised as three aspirated/unaspirated pairs — /ts tsʰ/, /tɕ tɕʰ/, /tʂ tʂʰ/
- Italian: /tʃ/ (ciao) and /dʒ/ (giorno, “day”), distributed by spelling rules
- Polish: four affricates — /tɕ dʑ tʂ dʐ/ — giving Polish one of the richer affricate inventories in European languages
At the other end of the typological range, many Polynesian languages have no affricates at all, and Classical Arabic lacks them in its native inventory.
Affricates in Japanese
Japanese is a strong case study for how affricates can emerge from allophonic conditioning rather than being stored as independent phonemes. The consonant /t/ in Japanese undergoes predictable changes depending on the following vowel: before /i/ it surfaces as [tɕ] (the chi in ち), and before /ɯ/ it surfaces as [ts] (the tsu in つ). Native speakers don’t perceive these as distinct sounds — they are all mentally /t/ — but phonetically they are affricates.
| Kana | Romaji | IPA | Example |
|---|---|---|---|
| ち | chi | [tɕi] | 力 (ちから, chikara, “power”) |
| つ | tsu | [tsɯ] | 月 (つき, tsuki, “moon”) |
| じ / ぢ | ji / di | [dʑi] | 時間 (じかん, jikan, “time”) |
The parallel voicing alternation affects /d/ as well: /d/ before /i/ surfaces as [dʑ] (ぢ/じ), and before /ɯ/ as [dz] (づ).
For learners whose L1 is English, the main challenge is [ts] at the start of a syllable. English phonotactics only permit /ts/ in syllable-final position (cats, bits), so the initial [ts] of つ (tsu) sits in an unfamiliar slot. English speakers frequently reduce it to “su” or “tu.” The fix is typically to practice the cluster explicitly — English already has the component sounds, they just don’t appear in that position.
Related Terms
See Also
Research / Sources
- Ladefoged, P., & Johnson, K. (2014). A Course in Phonetics (7th ed.). Cengage Learning.
- Vance, T. J. (2008). The Sounds of Japanese. Cambridge University Press.