•
Loading...
Checking authentication...
Learn linking techniques, assimilation, and elision to speak more naturally and fluently.
Explore our comprehensive pronunciation guides with audio and video examples.
Browse Pronunciation GuidesListen to a native English speaker in casual conversation, and you'll notice something curious: their speech flows like a continuous stream rather than discrete, separate words. "What are you doing?" doesn't sound like four distinct words but rather like "Whadda-ya-doin'?" This isn't lazy speech or poor articulation—it's connected speech, the natural phenomenon that makes English sound fluid and native-like.
For language learners, this presents a paradox. You've worked hard to pronounce each word correctly in isolation, but when native speakers talk, the words blur together in ways that seem to violate the pronunciation rules you've learned. The secret isn't in perfecting individual words but in understanding how words connect, blend, and transform when placed side by side. Welcome to the world of linking and connected speech.
Before diving into specific techniques, let's understand the fundamental principle: English speakers are physiologically lazy—in the most efficient way possible. The human mouth naturally seeks the path of least resistance when producing sounds. Moving your tongue, lips, and jaw requires muscular effort, and your mouth instinctively minimizes unnecessary movements.
When we speak in our native language, we're not thinking about individual phonemes (sound units). Instead, we're thinking in phrases and sentences, and our mouths develop efficient shortcuts. These shortcuts create linking, where sounds connect smoothly, and reduction, where sounds weaken or disappear.
This efficiency principle explains why non-native speakers who pronounce each word perfectly can still sound "foreign"—they're maintaining boundaries between words that native speakers naturally erase. Learning connected speech isn't about speaking "worse"; it's about speaking more naturally and efficiently.
The most fundamental type of linking occurs when a word ending in a consonant meets a word beginning with a vowel. In this case, the consonant sound slides directly into the vowel sound, creating a seamless connection that makes the phrase sound like a single word.
The Mechanics:
Consider "pick up." If you pronounce each word separately with pauses, it sounds choppy: "pick" [pause] "up." But native speakers connect them: "pi-kup," where the K sound links directly to the U sound. The consonant becomes the onset (beginning) of the next syllable.
Common Examples:
This pattern is so consistent that once you internalize it, it becomes automatic. Any time a consonant meets a vowel across word boundaries, link them.
Practice Technique: Write out phrases with consonant-to-vowel linking as if they were single words with hyphens showing the new syllable boundaries. Say "an apple" as "a-napple," "in an hour" as "i-na-nour." This visual reorganization helps your brain repattern the sounds.
When two vowel sounds meet across word boundaries, English inserts a tiny, almost imperceptible consonant sound to bridge them. This happens unconsciously, but it's crucial for natural-sounding speech. There are two primary types:
The /w/ Glide (for vowels produced with rounded lips):
When words ending in /u/, /oʊ/, or /aʊ/ sounds meet words beginning with vowels, a subtle W sound appears:
The W sound is barely there—it's not a full, strong W like in "water." It's a glide, a momentary transition that keeps the vowels from colliding awkwardly.
The /j/ Glide (for front vowels):
When words ending in /i/, /eɪ/, or /aɪ/ sounds meet words beginning with vowels, a subtle Y sound (represented as /j/ in phonetics) appears:
Again, this Y sound is subtle—you're not adding a full "yuh" sound but rather allowing your tongue to glide naturally from one vowel position to the next.
Why It Matters: Without these glide sounds, vowel-to-vowel transitions sound abrupt and foreign. Compare "I am" with a glottal stop (a tiny pause) versus "I-yam" with smooth linking. The linked version sounds natural; the separated version sounds like a robot or someone speaking very carefully.
When words ending in consonants meet words beginning with consonants, linking becomes more complex. Several patterns emerge:
Identical Consonants (Gemination):
When the same consonant appears at the end of one word and the beginning of the next, we don't pronounce it twice. Instead, we hold the consonant slightly longer:
The colon (:) represents the elongation—you're not saying two separate consonants but holding one slightly longer than normal.
Stop Consonants Before Other Consonants:
Stop consonants (P, B, T, D, K, G) involve completely stopping airflow, then releasing it. When a stop consonant ends a word and another consonant begins the next word, we often don't release the first stop audibly. This creates an unreleased or held stop:
This creates a subtle pause or tension point between the words, but you don't add an extra vowel sound or fully release the consonant.
Now we enter more advanced territory: assimilation, where sounds actually change to become more similar to their neighbors. This is the ultimate expression of articulatory laziness—sounds morph to minimize mouth movement.
Place Assimilation (The Most Common Type):
Sounds shift their place of articulation (where in the mouth they're produced) to match nearby sounds. The most frequent example involves alveolar sounds (/t/, /d/, /n/) before bilabial or velar sounds:
Why does this happen? Pronouncing N requires your tongue to touch the alveolar ridge (behind your upper teeth). But P, B, and M are made with your lips. Rather than moving from tongue-position to lip-position, your mouth compromises by making everything lip-based: N becomes M.
Similarly, before K or G (made at the back of the mouth), N shifts to the NG sound (as in "sing"), which is also made at the back of the mouth.
Yod Coalescence (The "Got You" Phenomenon):
When certain consonants meet the Y sound (/j/), they fuse into new sounds:
This assimilation is so ingrained in casual English that saying "did you" without the J sound ("did-you" with clear separation) sounds formal or even stilted.
Voicing Assimilation:
Less common in English than in some languages, but still relevant: voiced consonants can devoice (lose their voicing) before voiceless consonants, and vice versa:
| Assimilation Type | Example Phrase | Phonetic Result | Why It Happens |
|---|---|---|---|
| Place (N→M) | ten people | "tem people" | Matching articulation location |
| Place (N→NG) | ten cars | "teng cars" | Matching articulation location |
| Yod (T+Y→CH) | can't you | "can'tcha" | Consonant + glide fusion |
| Yod (D+Y→J) | did you | "didja" | Consonant + glide fusion |
| Voicing (V→F) | have to | "hafta" | Voiceless context influence |
Elision takes efficiency to the next level: sounds vanish entirely. This happens most frequently with sounds that are difficult to articulate in particular contexts.
T and D Deletion:
The most common elision in English involves T and D sounds disappearing between consonants or at word endings:
T and D are particularly vulnerable because they're alveolar stops—they require precise tongue placement. When surrounded by other consonants, this precise placement becomes difficult, so the sound drops out.
H-Dropping in Function Words:
In casual speech, H often disappears from unstressed pronouns and auxiliary verbs:
This H-dropping is standard in connected speech, not "sloppy" pronunciation. However, it only occurs in unstressed contexts—"HIM" when emphasized retains its H.
Schwa Insertion and Deletion:
The schwa (ə), English's most common vowel sound (the "uh" in "about"), frequently disappears from unstressed syllables:
Conversely, sometimes schwas are inserted to break up difficult consonant clusters:
The flip side of elision is intrusion—sounds that appear where they don't "belong" according to spelling. We've already discussed W and Y intrusion in vowel-to-vowel linking. Another common type is:
R-Intrusion (in non-rhotic accents):
In British English and some other accents where R at the end of words isn't pronounced (non-rhotic accents), an R sound appears between certain vowels:
This happens because words like "law" historically had an R (from Old English), and the linking pattern persists even after the R was dropped in isolation. It's called "linking R" when the R exists in spelling ("far away" → "far-r-away") and "intrusive R" when it doesn't ("law and" → "law-r-and").
Let's analyze real phrases showing multiple connected speech features simultaneously:
Example 1: "What are you going to do about it?"
Full connected speech form: "Whadda-ya-gonna-do-wa-bou-dit?"
Example 2: "I should have asked him yesterday."
Full connected speech form: "I-shda-sk-im-yesterday"
Example 3: "Did you see the news last night?"
Full connected speech form: "Didja-see-thi-news-las-night?"
Exercise 1: Consonant-to-Vowel Drilling
Practice these phrases, focusing on smooth linking without pauses:
Exercise 2: "Got You" Transformations
Practice yod coalescence by transforming these formal phrases into casual forms:
Exercise 3: Function Word Reduction
Practice reducing and linking function words naturally:
Exercise 4: Minimal Pairs
Compare careful speech (with boundaries) versus connected speech (with linking):
| Careful Speech | Connected Speech |
|---|---|
| pick. up. | pi-kup |
| see. you. | see-yuh |
| go. away. | go-waway |
| big. game. | bi-g:ame |
| did. you. | di-juh |
| ten. people. | tem-people |
Record yourself saying both versions and compare. The connected speech version should feel smoother and faster.
Exercise 5: Shadowing Native Speakers
Find recordings of natural conversation (podcasts, TV shows, interviews—not newscasters, who speak more carefully). Listen to short segments (5-10 seconds), then immediately repeat exactly what you heard, mimicking the rhythm, linking, and reductions. Don't look at transcripts initially—train your ear first.
Pitfall 1: Over-articulating
Learners often pronounce every word clearly and separately, creating unnatural pauses. This sounds robotic. Instead, think in phrases, not words. Connect everything within a thought group.
Pitfall 2: Under-articulating
Conversely, trying too hard to sound casual can lead to mushy, unclear speech. The goal is efficiency, not sloppiness. Key content words should remain intelligible.
Pitfall 3: Applying Linking Inconsistently
Don't pick and choose when to link. Native speakers link automatically and consistently. If a consonant meets a vowel, link them. If a word ends and begins with the same consonant, don't double it.
Pitfall 4: Ignoring Stress Patterns
Connected speech features like reduction and elision primarily affect unstressed syllables and function words. Content words (nouns, main verbs, adjectives, adverbs) remain relatively clear. Don't reduce everything uniformly.
Pitfall 5: Formal vs. Casual Context Confusion
Connected speech exists on a spectrum. Public speeches, formal presentations, and careful reading involve less reduction and linking than casual conversation. Be aware of context and adjust accordingly.
It's worth noting that connected speech patterns vary by region, social group, and speaking style. Some assimilations and reductions are more common in certain English varieties:
Additionally, speakers adjust their connected speech usage based on formality, audience, and situation. You'll use more linking and reduction with friends than in a job interview.
Phase 1: Awareness (Weeks 1-2)
Focus on hearing connected speech. Watch TV shows or movies with subtitles, noting when what you hear doesn't match what's written. Identify linking, assimilation, and reduction instances. Make a list of common patterns you notice.
Phase 2: Mimicry (Weeks 3-4)
Practice shadowing exercises daily. Choose speakers you want to sound like and imitate their connected speech precisely. Don't worry about understanding everything—focus on matching the sound flow.
Phase 3: Active Practice (Weeks 5-8)
Consciously apply linking and reduction in your own speech. Start with high-frequency phrases ("going to," "want to," "did you," etc.) until they become automatic. Gradually expand to longer utterances.
Phase 4: Integration (Ongoing)
Make connected speech your default. Record yourself weekly and listen critically: Are you maintaining word boundaries unnecessarily? Are you linking smoothly? Are function words reduced appropriately?
Learning connected speech transforms your English from textbook-correct to genuinely fluent. It's not about abandoning proper pronunciation but about adapting it to the natural flow of real communication. When you master linking, assimilation, elision, and reduction, you don't just sound more native—you think more like a native speaker, processing language in phrases and rhythm rather than isolated words.
The journey from "What. Are. You. Doing?" to "Whadda-ya-doin'?" isn't about lowering standards—it's about achieving true fluency. Your mouth learns to dance with the language rather than march through it.