Ever tried to say the word loch like a true Scot, with that gravelly catch in the back of your throat? For most English speakers, the attempt ends in one of two ways: it either comes out as “lock”, or it feels like you’re awkwardly trying to clear your throat. You might chalk it up to a personal failing, a sound you just can’t make. But what if I told you it’s not you, it’s the English language itself?
This struggle highlights a fascinating corner of linguistics: the world of marginal phonemes. These are the sounds that are technically *in* a language but live on the absolute fringes, like guests at a party who never take their coats off. They haven’t fully integrated, and they reveal a lot about how sound systems work, evolve, and sometimes, reject newcomers. They are, in essence, failed phonemes—sounds that died at birth.
Before we dive into the failures, let’s remember the stars of the show. A phoneme is the smallest unit of sound that can distinguish one word from another. They are the fundamental building blocks of a language’s soundscape.
Think of the /p/ and /b/ sounds in English. The only difference between “pat” and “bat” is that single phoneme, and it completely changes the meaning. A language’s phonemic inventory is the set of all such contrastive sounds that its speakers intuitively use. For English, this includes sounds like /t/, /d/, /k/, /g/, /s/, /z/, and a couple dozen more. These are the A-team, the fully-fledged members of the club.
Marginal phonemes, on the other hand, are the B-team. They show up, but only in very specific, limited circumstances.
So, which sounds are lingering on the periphery of English? The list is short but revealing, and almost all of them come from other languages.
The infamous sound in loch is a voiceless velar fricative, represented in the International Phonetic Alphabet (IPA) as /x/. It’s the sound in the German Bach, the Spanish jota, and the Russian хорошо. In English, it appears almost exclusively in loanwords from Scottish Gaelic (loch), German (Bach), Yiddish (chutzpah), or Hebrew (Chanukah).
Because it’s not a native English sound, most speakers substitute it with the closest “legal” sound in their phonological toolbox: /k/. So, loch becomes “lock” and Bach becomes “bock.” The fact that we have a widespread, acceptable substitution is the biggest clue that /x/ isn’t a core part of the system. There are no minimal pairs in English that hinge on the /k/ vs. /x/ distinction, except for these few, often hyper-corrected, loanwords.
Have you ever ordered a croissant or talked about your je ne sais quoi? If so, you’ve encountered French nasal vowels. Sounds like the “an” in encore or the “on” in bon vivant are formed by passing air through both the mouth and the nose.
While some English speakers who know French might produce them accurately, the vast majority of us “nativize” them. We pronounce the vowel and then add a distinct /n/ or /m/ sound at the end. Encore becomes “ON-core.” This process, called denasalization, shows that these nasal vowels have no status as independent phonemes in English. They only exist in unassimilated French phrases, clinging to their foreign identity.
The glottal stop is the little catch in your throat in the middle of “uh-oh!”. It’s also the sound that famously replaces the /t/ in words like “butter” (bu’er) and “bottle” (bo’le) in many British and some American dialects.
So, is it a phoneme? In most English dialects, the answer is no. When a Cockney speaker says “bu’er”, they are not using a different word from someone who says “butter.” The glottal stop is an allophone—a different pronunciation of the same phoneme, /t/. Because it doesn’t create a new meaning, it doesn’t get phoneme status. Its only independent appearance is in “uh-oh!”, which is more of an interjection than a standard word, further cementing its marginal nature.
A sound doesn’t just show up and get a job as a phoneme. It has to audition, and most fail. Here are the main reasons why.
Can a marginal phoneme ever go mainstream? Absolutely! In fact, one of English’s now-standard sounds is a poster child for this very process.
Meet /ʒ/, the voiced palato-alveolar fricative. It’s the sound in measure, vision, and beige.
This sound did not exist in Old English. It first entered the language via a massive wave of French loanwords after the Norman Conquest in 1066 (e.g., rouge, genre). For a long time, it was a marginal, foreign sound. However, a funny thing happened. A natural sound change called palatalization started occurring within English itself, where /zj/ combinations morphed into /ʒ/ (as in vision, from Latin *visio*). As the sound started appearing in more and more words—both borrowed and nativized—it became a stable, productive part of the English phonological system. It had graduated from a marginal guest to a full-fledged citizen.
The story of /ʒ/ shows that integration is possible, but it requires the right conditions: a high volume of borrowing and, crucially, a natural pathway for the sound to emerge in the language’s own internal processes.
So, the next time you hesitate over Chanukah or stumble on croissant, don’t sweat it. You’re not failing a pronunciation test; you’re running into the invisible walls of your language’s sound system. These “failed phonemes” are the fascinating ghosts in the machine, reminding us that language is a living, breathing entity, constantly deciding which sounds are worth keeping and which are just passing through.
Contrary to Hollywood depictions, lip-reading is less like a superpower and more like a high-stakes…
Which came first: the editor or the edit? The answer reveals a fascinating linguistic process…
Ever wonder why "Grandma's slow-cooked apple pie" sounds more appealing than just "apple pie"? The…
Ever wonder why people in isolated places like an Appalachian hollow develop such a unique…
Ever wonder why scientists use a "dead" language to name living things? Scientific Latin is…
Unlike English, the Irish language doesn't have a single verb for "to have." Instead, to…
This website uses cookies.