Fricatives
Chapter Objectives
By the end of this chapter, you will be able to:
- Explain how frication is generated in articulatory terms.
- Distinguish sibilant and nonsibilant fricatives.
- Classify major fricatives by place and voicing.
- Use auditory cues and an introductory spectrogram preview to identify frication.
- Compare nearby fricative places across languages without claiming identical realization.
- Read and transcribe a controlled set of fricative symbols accurately.
- Avoid assuming that every voiced fricative sustains voicing identically in every context.
An Observable Problem
Say the English words sip and ship slowly, listening only to the first consonant. Both involve a narrow channel that forces air through with audible noise — but the noise itself sounds different, and if you watch a friend's lips you may notice their tongue is in a different place for each. Now say fin and thin. These two also involve turbulent airflow, but the noise is much quieter, almost a hiss without an edge to it, and the place of articulation has moved again.
All four of these consonants share one property: none of them involves a complete closure of the vocal tract, the way the stops in Chapter 9 did. Instead, the articulators come close enough together to create audible friction. This chapter asks a more precise question than "is there noise?" It asks where that noise is produced, how loud and sharp it is, and whether the vocal folds are vibrating at the same time.
- Before reading further, say sip, ship, fin, and thin aloud and rank the four initial consonants from "sharpest, most hissing noise" to "quietest noise." Write down your ranking; you will return to it later in the chapter.
How Frication Is Generated
Purpose: connect the auditory impression of "noisy" consonants to a precise articulatory mechanism.
A fricative is a consonant produced by narrowing the vocal tract enough to force air through a small channel, creating audible turbulence, without ever reaching a complete closure. This distinguishes fricatives from the stops studied in Chapter 9, where the airstream is briefly blocked entirely, and from approximants, previewed in Chapter 8 and developed fully in Chapter 12, where the articulators approach one another without creating turbulent noise.
The turbulence itself, not any single articulator, is the defining acoustic signature of a fricative. Air molecules moving through a narrow constriction at sufficient speed collide unpredictably, producing aperiodic, noise-like energy that a listener perceives as hissing, buzzing, or rasping depending on the exact shape and location of the channel.
- [fɪn]fin — turbulence generated between the lower lip and upper teeth
- [sɪp]sip — turbulence generated between the tongue tip or blade and the alveolar ridge, directed at the teeth
A useful working test: if you can sustain a consonant indefinitely on a single breath while air continues to escape audibly, and the sound is not a vowel, nasal, or approximant, it is very likely a fricative. Stops cannot be sustained this way because the closure stops the airflow; fricatives can be, because the constriction never fully closes.
Sibilant and Nonsibilant Fricatives
Purpose: distinguish two acoustically and articulatorily distinct classes of fricative noise.
A sibilant fricative is produced with a constriction shaped so that the airstream is directed at the edge of the teeth, producing comparatively loud, high-frequency, sharply audible noise. A nonsibilant fricative lacks this tooth-directed channeling and produces quieter, less sharply defined noise.
- [s], [z], [ʃ], [ʒ]sibilants — comparatively loud, high-frequency frication
- [f], [v], [θ], [ð]nonsibilants — comparatively quiet, diffuse frication
Sibilance is a property of the acoustic output, not a separate place of articulation. Sibilant fricatives occur at several different places, including alveolar and postalveolar, while sharing the tooth-directed channeling that produces their characteristic loudness.
On a spectrogram, introduced fully in Chapter 28, sibilant fricatives typically show dense, high-amplitude noise energy concentrated above roughly 3500 Hz, while nonsibilant fricatives such as [f] and [θ] show much weaker, more spread-out energy that can be difficult to distinguish visually from background noise. This preview is meant only to connect the auditory contrast to a visible correlate; full spectrogram interpretation is developed later in the course.
Fricative Places of Articulation
Purpose: organize the fricative series by place, recycling the place categories introduced in Chapter 7.
Fricatives occur at nearly every place of articulation introduced in Chapter 7. The table below presents a controlled set of voiceless and voiced fricative symbols organized by place, sufficient for the classification and transcription work in this chapter.
| Place | Voiceless | Voiced | Sibilant? |
|---|---|---|---|
| Labiodental | f | v | No |
| Dental | θ | ð | No |
| Alveolar | s | z | Yes |
| Postalveolar | ʃ | ʒ | Yes |
| Glottal | h | — | No |
Additional fricative places, including bilabial, alveolo-palatal, retroflex, velar, uvular, pharyngeal or epiglottal, and additional voiced-voiceless pairs, are receptive at this stage and will become productive as scheduled in later chapters and the Phonetics Progression Matrix.
A note on [h]
The symbol [h] is conventionally classed with the fricatives, but it is somewhat unusual: it has no fixed oral constriction location and is better described as audible turbulence generated at or near the glottis, with the oral cavity shaped by whatever vowel follows. Treat [h] as a glottal fricative for classification purposes, while noting that its place is less precisely localized than the other fricatives in the table.
Error: Assuming that ⟨th⟩ always represents the same sound, since it is spelled with the same two letters in thin and this.
Correction: English ⟨th⟩ represents two different dental fricative phonemes depending on the word: voiceless [θ] in thin [θɪn], and voiced [ð] in this [ðɪs]. Spelling does not distinguish them; pronunciation does.
Voicing Pairs and Their Limits
Purpose: apply the voicing concept from Chapter 6 to fricatives, and qualify it with the evidence fricatives actually provide.
Many fricative places have both a voiceless and a voiced member produced at the same location, as the table above shows. It is tempting to treat each pair as differing only in whether the vocal folds vibrate, with everything else held constant. This is a useful starting approximation, but it is not the whole picture.
Sustained turbulent airflow and sustained vocal-fold vibration compete aerodynamically: vibration requires airflow that maintains pressure below the glottis, while frication requires airflow forced through a narrow channel above it. As a result, voiced fricatives in many languages, including English, frequently show partial devoicing in certain positions, particularly at the end of an utterance, even though they remain phonemically voiced.
- /hæz/has — phonemically ends in /z/; the final fricative is often only partially voiced in careful citation speech
The claim that a fricative is "voiced" is a phonemic claim about category membership, not a promise that vocal-fold vibration is audible for the sound's entire duration in every token. Treat voicing labels for fricatives as the contrastive category established by the language's phoneme system, while recognizing that the phonetic realization can vary by position, speaker, and rate. This distinction returns when you study connected speech in Chapter 23 and acoustic voice onset evidence in Chapter 28.
Across Languages: Comparing Nearby Fricative Places
Purpose: show that a shared symbol does not guarantee an identical realization, using verified cross-language examples.
The IPA symbol [s] represents an alveolar sibilant fricative wherever it is used, but languages differ in exactly how that constriction is shaped, and some languages contrast more than one sibilant place where English has only one.
- [s] — Spanish (Castilian and Latin American varieties), as in sopa [ˈsopa]soup — alveolar sibilant, contrasting with no second sibilant place in most varieties
- [ɕ] — Mandarin Chinese, as in xià [ɕja˥˩]down — alveolo-palatal sibilant, a distinct phoneme from the alveolar and retroflex sibilants Mandarin also contrasts
Mandarin's three-way sibilant contrast (alveolar, alveolo-palatal, and retroflex) shows that the single alveolar-versus-postalveolar distinction familiar from English is not a universal limit on how many sibilant places a language can use contrastively. The alveolo-palatal and retroflex symbols remain receptive at this stage; they are introduced for productive use according to the progression matrix.
Within English itself, the realization of /s/ varies somewhat by speaker and regional accent in its precise tongue shape and acoustic quality, even though it remains the same phoneme. This kind of fine phonetic variation is normal and does not by itself indicate a different phoneme or a speech difference requiring correction.
Guided Observation
Complete this short self-observation task before moving to the dataset analysis. No audio recording is required — you are listening to and feeling your own speech.
Say each pair below slowly and notice two things: whether you can feel vocal-fold vibration by lightly touching your throat, and whether the noise sounds sharp and hissing or quiet and diffuse.
- fan vs. van
- sue vs. zoo
- thigh vs. thy
For each pair, write one sentence stating which member you judge to be voiced and which sibilant class, if any, each consonant belongs to. Compare your judgments with the ranking you wrote in the opening scenario.
Listening For Frication
Purpose: identify the perceptual cues that distinguish fricative place and class before relying on spelling or memory.
- Overall loudness and sharpness of the noise: sibilants are louder and sharper than nonsibilants.
- Pitch quality of the noise itself: postalveolar frication tends to sound lower-pitched and "fuller" than alveolar frication.
- Presence or absence of voicing throughout the consonant, checked by lightly touching the throat.
- Duration: fricatives, like other continuants, can be sustained and compared at different lengths, unlike stops.
Dataset Analysis Lab
Purpose: apply place, voicing, and sibilance classification to a small, self-contained dataset.
The forms below are verified phonetic transcriptions of English words, General American reference variety, spoken in careful citation style. Use only this dataset to answer the questions that follow — do not rely on outside knowledge of English spelling.
| Orthographic form | Phonetic transcription | Gloss |
|---|---|---|
| ⟨fan⟩ | [fæn] | a device that moves air |
| ⟨van⟩ | [væn] | a type of vehicle |
| ⟨sip⟩ | [sɪp] | to drink in small amounts |
| ⟨zip⟩ | [zɪp] | to fasten with a zipper |
| ⟨ship⟩ | [ʃɪp] | a large seafaring vessel |
| ⟨thigh⟩ | [θaɪ] | part of the leg |
| ⟨thy⟩ | [ðaɪ] | archaic for "your" |
| ⟨measure⟩ | [ˈmɛʒɚ] | to determine a quantity (medial fricative shown for contrast) |
Source: forms checked against a standard American English pronouncing reference for General American citation pronunciation. The transcription level is broad-to-moderate phonetic detail, sufficient to show the contrasts under analysis.
Step 1 — Classify by place and voicing
For each fricative in the dataset, state its place of articulation and whether it is voiced or voiceless. Use the table from this chapter's Fricative Places section as your reference.
Step 2 — Classify by sibilance
Sort the eight fricatives in the dataset into two groups: sibilant and nonsibilant. State the articulatory or acoustic property that justifies each grouping.
Step 3 — Test for contrast
Using the contrast test from Chapter 5, determine whether fan [fæn] and van [væn] provide evidence that [f] and [v] are separate phonemes of English. Justify your answer and state the correct delimiter for your conclusion.
Step 4 — Write the analysis
In two to three sentences, summarize the place, voicing, and sibilance properties of the fricatives in this dataset, citing at least two specific forms as evidence.
Practice — Transcription: Transcribe the first consonant of each word below in square brackets, using only symbols introduced in this chapter's table.
- vote
- think (the consonant in thin, not in this)
- show
- zest
Common Mistakes
Error: Treating "sibilant" as a place of articulation rather than a property of the resulting noise.
Correction: Sibilance describes the acoustic and articulatory channeling that produces loud, high-frequency frication; it can occur at more than one place, as shown by alveolar [s] and postalveolar [ʃ], both sibilants at different places.
Error: Assuming a voiced fricative is voiced for its entire duration in every context, and treating any partial devoicing as a pronunciation error.
Correction: Voicing labels for fricatives identify a contrastive phonemic category. Phonetic realization can include partial devoicing in predictable positions, such as utterance-finally, without changing the phoneme's identity.
Error: Assuming that every language with an [s]-like sound has only one sibilant phoneme, the way many English varieties do.
Correction: Some languages, including Mandarin Chinese, contrast more than one sibilant place. The number of sibilant contrasts is a property of each language's phoneme system, not a universal fact about sibilants.
Chapter Review
Part A — Definitions
1. In your own words, define fricative and explain how it differs from a stop and from an approximant.
2. Define sibilant and give two English examples not used elsewhere in this chapter's main discussion.
Part B — Classification
1. Classify [ʒ] by place, voicing, and sibilance.
2. Explain why [h] is conventionally classified with the fricatives even though its place of articulation is less fixed than that of the other fricatives in this chapter.
Part C — Voicing in Context
1. Explain why a fricative described as "voiced" might not show audible vocal-fold vibration throughout its entire duration in every context.
Part D — Cross-Language Comparison
1. Using the Mandarin and Spanish examples from this chapter, explain why the existence of the symbol [s] in two languages does not guarantee that both languages have the same number of sibilant phonemes.
Repaso acumulativo / Cumulative Review
1. Using the manner-classification framework from Chapter 8, explain why affricates, studied in Chapter 9, require reference to frication even though they are not classified as fricatives.
2. Using the phoneme-allophone framework from Chapter 5, state whether thigh [θaɪ] and thy [ðaɪ] form a minimal pair, and explain what this shows about the phonemic status of [θ] and [ð] in English.
Chapter Performance Task
Performance Task
Building a Fricative Classification Chart
Using the controlled symbol set introduced in this chapter:
- Collect at least six English words illustrating at least four different fricatives, written in ordinary spelling.
- Transcribe the relevant consonant in each word phonetically, in square brackets.
- Build a small table classifying each fricative by place, voicing, and sibilance.
- Identify one minimal or near-minimal pair from your collected words and state what it shows about contrast in English.
- Identify one piece of additional evidence, such as a recording or a second speaker's pronunciation, that would strengthen your classification.
Answer Key (click to expand)
Dataset Analysis Lab
Step 1: [f] labiodental voiceless; [v] labiodental voiced; [s] alveolar voiceless; [z] alveolar voiced; [ʃ] postalveolar voiceless; [θ] dental voiceless; [ð] dental voiced; [ʒ] postalveolar voiced.
Step 2: Sibilants: [s], [z], [ʃ], [ʒ], produced with the airstream channeled toward the teeth, yielding loud, high-frequency noise. Nonsibilants: [f], [v], [θ], [ð], lacking this tooth-directed channeling and producing quieter, more diffuse noise.
Step 3: Fan [fæn] and van [væn] form a minimal pair: they differ in meaning and in exactly one sound, the initial consonant, with the rest of the word identical. This is evidence that [f] and [v] belong to separate phonemes, written /f/ and /v/.
Step 4 (model answer): The dataset contains four places of articulation — labiodental, dental, alveolar, and postalveolar — each with a voiceless and voiced member, as shown by the [s]/[z] and [θ]/[ð] pairs. The alveolar and postalveolar fricatives are sibilant, while the labiodental and dental fricatives are nonsibilant, a distinction supported by the comparatively sharp, loud quality of [s] and [ʃ] relative to [f] and [θ]. Other well-reasoned answers that correctly apply the classification criteria are acceptable.
Practice — Transcription
1. [v] 2. [θ] 3. [ʃ] 4. [z]
Chapter Review — Part A (model answers)
1. A fricative is produced by narrowing the vocal tract enough to create audible turbulence without a complete closure; this distinguishes it from a stop, which involves complete closure, and from an approximant, which involves a narrowing too open to generate turbulent noise. 2. A sibilant is a fricative whose airstream is channeled toward the teeth, producing comparatively loud, high-frequency noise; answers will vary, e.g. zoo [zu] and shoe [ʃu].
Chapter Review — Part B
1. [ʒ] is postalveolar, voiced, and sibilant. 2. [h] is classed with the fricatives because it involves audible turbulence generated near the glottis, but its oral place is not fixed because the oral cavity simply takes the shape of the following vowel rather than forming an independent oral constriction.
Chapter Review — Part C
1. Sustained frication and sustained vocal-fold vibration compete aerodynamically, so a phonemically voiced fricative can show partial or complete devoicing in certain contexts, such as utterance-finally, without losing its phonemic status as voiced.
Chapter Review — Part D
1. The symbol [s] represents the same articulatory and acoustic target, alveolar sibilant frication, in both languages, but the number of sibilant phonemes a language contrasts is a separate, language-specific fact. Spanish typically contrasts one sibilant place in most varieties, while Mandarin contrasts three (alveolar, alveolo-palatal, and retroflex), so sharing a symbol does not mean the two languages organize their sibilant inventories the same way.
Cumulative Review
1. Affricates combine a stop closure with a fricative release at the same or a closely related place, so describing the release phase of an affricate requires the same frication concepts introduced in this chapter, even though affricates are classified as their own manner category rather than as fricatives. 2. Yes, thigh and thy form a minimal pair: they differ in meaning and in exactly one sound in the same position, with the rest of the word identical. This shows that [θ] and [ð] belong to separate phonemes in English, /θ/ and /ð/, rather than functioning as allophones of a single phoneme.
Looking Ahead
What You Can Now Do
You can now explain how frication is generated, distinguish sibilant from nonsibilant fricatives, classify fricatives by place and voicing, and recognize that a shared symbol does not guarantee an identical realization or an identical inventory across languages. These skills extend the place-and-manner framework from Chapters 7 and 8 and prepare you for the remaining consonant classes.