Coarticulation is the way your mouth blends sounds together during speech, adjusting how you produce one sound based on the sounds that come before and after it. Rather than pronouncing each sound in isolation like beads on a string, your tongue, lips, and jaw are constantly overlapping their movements, shaping the current sound while already preparing for the next one. This blending is so fundamental to how humans talk that speech without it would sound robotic and unnaturally slow.
How Coarticulation Works
When you speak, your vocal tract (the space from your vocal cords to your lips) is controlled by a complex set of muscles moving the tongue, jaw, lips, and soft palate. These structures are physical objects with mass and momentum. They can’t teleport from one position to another between sounds, so their movements naturally overlap. Your tongue might already be rising toward the roof of your mouth for an upcoming “t” while your lips are still finishing a rounded vowel.
Some of this blending is purely mechanical. The jaw, for instance, serves as a platform for the tongue and lower lip. Its physical properties, including muscle mechanics and the geometry of the jawbone itself, create movement patterns that overlap regardless of any conscious planning. But coarticulation isn’t entirely automatic. Research in neuroscience has shown that some coarticulatory effects are centrally planned by the brain, while others emerge from the physics of the muscles and joints involved. The brain sends motor commands that already account for upcoming sounds, building overlap into the plan from the start.
Interestingly, coarticulation patterns vary between languages. While some blending is universal because all humans share the same basic vocal anatomy, different languages develop their own specific patterns of how sounds influence each other. A speaker of Japanese and a speaker of English will coarticulate the same sequence of sounds somewhat differently, reflecting the phonetic habits of their native language.
Anticipatory vs. Carryover Coarticulation
Linguists describe two directions of coarticulation based on which sound is influencing which.
Anticipatory coarticulation happens when a sound is shaped by the one that follows it. Your mouth essentially “looks ahead” and starts preparing for the next sound early. A classic example: say the words “key” and “cool” out loud. The “k” in each word is produced in a noticeably different position. In “key,” your tongue makes the “k” contact farther forward in your mouth because it’s already anticipating the front vowel “ee.” In “cool,” the “k” is produced farther back because the following “oo” is a back vowel. You can also feel your lips rounding during the “g” in “goon” because they’re getting ready for the rounded “oo” that comes next.
Carryover coarticulation (also called perseverative coarticulation) works in the opposite direction. Here, a sound’s influence lingers and affects the sound that follows it. In the word “boots,” for example, the lip rounding required for the “oo” vowel carries over onto the “t” and “s” at the end. Your lips don’t snap back to a neutral position the instant the vowel ends; that rounding persists through the final consonants. Carryover effects are generally considered a consequence of articulatory inertia, meaning the speech organs simply can’t reset instantaneously. Though some researchers argue that even carryover coarticulation involves a degree of motor planning, the physical momentum of the articulators is the primary driver.
Vowel Nasalization as an Example
One of the most noticeable coarticulation effects in American English is vowel nasalization. When a vowel sits next to a nasal consonant like “m,” “n,” or “ng,” the vowel itself takes on a nasal quality because the soft palate (the fleshy part at the back of the roof of your mouth) begins lowering early, letting air flow through the nose before the nasal consonant officially begins.
You can hear this in a word like “bomb.” The vowel in the middle picks up nasalization from the surrounding “b” and “m” sounds. Researchers measure this effect by comparing the acoustic energy of the vowel’s natural resonance against the energy coming through the nasal cavity. The degree of nasalization varies depending on factors like how much emphasis a speaker places on the word and where it falls in a sentence. Stressed words and words at the boundaries of phrases tend to show different nasalization patterns than unstressed words buried in the middle of a sentence. Each speaker also has their own individual nasalization habits, making this a rich area of variation in everyday English.
What Coarticulation Sounds Like on a Spectrogram
If you could visualize speech on a spectrogram (a graph showing sound frequencies over time), coarticulation appears as smooth, curving transitions between sounds rather than sharp, sudden jumps. The dark bands on a spectrogram, called formants, represent the resonant frequencies of the vocal tract. When a consonant is produced between two vowels, these formant bands bend and shift in predictable ways.
For an alveolar consonant (one made with the tongue near the ridge behind the upper teeth, like “t” or “d”) placed between two vowels, the first formant drops by nearly 200 Hz as the tongue moves into position for the consonant, while the second and third formants rise by 300 Hz or more. After the consonant releases, these formants reverse course and settle into the pattern of the following vowel. These transitions aren’t random noise. They carry crucial information about which consonant is being produced and which vowel is coming next. The smooth bending of formants is, in fact, one of the main acoustic signatures of coarticulation.
Why Coarticulation Helps Listeners
Coarticulation might seem like sloppy pronunciation, but it actually makes speech easier to understand, not harder. Because each sound carries traces of its neighbors, listeners get advance acoustic information about what’s coming next. When you hear the beginning of a word, the coarticulatory cues baked into the first sound are already hinting at the second and third sounds. Your brain uses these cues to predict and identify upcoming sounds faster than it could if each sound were produced in isolation.
Listeners learn these patterns through experience with their native language. Since coarticulation patterns are partly language-specific, understanding a new language involves learning not just its sounds but also the way those sounds blend together. This is one reason why foreign-accented speech can be harder to process: the coarticulatory patterns don’t match what the listener’s brain expects.
Coarticulation in Speech Disorders
When the brain’s ability to plan speech movements breaks down, coarticulation is one of the first things affected. In acquired apraxia of speech, a neurological condition that impairs the planning and programming of speech movements, people struggle to smoothly sequence sounds together. Unlike dysarthria, which involves muscle weakness, apraxia leaves the muscles themselves intact. The problem is in the motor planning: the brain has difficulty organizing the overlapping movements that coarticulation requires.
People with apraxia typically find consonant clusters (like the “sk” in “skill”) much harder than single consonants (like the “s” in “sill”), precisely because clusters demand tighter coordination and more overlapping movements. Similarly, longer or more complex vowel sounds are harder than short, simple ones. Speech therapy for apraxia often uses rhythm and prosody-based approaches, training patients to re-establish the natural timing and flow that make coarticulation possible. These techniques aim to rebuild the smooth, overlapping movement patterns that healthy speakers produce without thinking.
Coarticulation sits at the intersection of physics and neuroscience. It exists because the human vocal tract is a physical system with real mass and momentum, and because the brain has evolved to exploit that physics, planning speech movements that deliberately overlap to produce fast, fluid, intelligible communication.