What is the origin of the pronunciation of words like Leicester, Gloucester, Worcestershire? Presumably, the spelling predates the pronunciation but what is the history here? What language do the words originate from?

Is this just simple elision or is there a more specific, shared origin of these words and the process that changed them?

  • 2
    It might be interesting to contrast the Lancaster set with the Manchester set with the Leicester set.
    – tchrist
    Sep 19, 2014 at 13:43
  • 4
    Actually, these are evidence that the spelling predates the pronunciation. I.e. when the spelling of these placenames became fossilized, they were pronounced with many more syllables than people give them nowadays.
    – Marthaª
    Sep 19, 2014 at 17:00
  • And, conversely, why is Cirencester pronounced siren-sester and are there any other exceptions to the "rule" that -cester is pronounced "stuh". Sep 19, 2014 at 18:35
  • 1
    @David: I don't believe it 8is (by the locals); more like "Sissester". Sep 19, 2014 at 20:25
  • @TimLymington Wikipedia disagrees, though without citation. But even "Sissester" would go against the pattern of all the other -cesters, which would suggest "Sirenster". Sep 19, 2014 at 21:01

4 Answers 4


These are all quite old names with many syllables—an ideal place for ellipsis, elision, and contraction.

The -cester bit is from the Old English word ceaster, which in itself is borrowed from Latin castrum. It is thus cognate to castle. (The palatalisation of the initial /k/ to /tʃ/ is common in Old English and was more widespread in some dialects than in others, so at some point, there were probably two variants around: caster /kastər/ and ceaster /tʃastər/.)

In Old English, ceaster referred to a Roman town or settlement (in England). As tchrist notes in his comment, some other places that have this element in their names include Lancaster, Doncaster, and Manchester.

The first part of names that have -cester tend to be inherited place names of either Anglo-Saxon or (even earlier) British Celtic stock, often names of rivers or other local toponyms.

Going from Old English down through the centuries, the cities the first element of whose names ended in a consonant (like Lan-, Don-, Man-) usually kept a fuller version of the word ceaster, retaining the initial consonant. The ones whose first element ended in a vowel, on the other hand, ended up having the initial /tʃ/ (or /k/, as in Lancaster and Doncaster) in an intravocalic position, where it was quite likely to be weakened—something that happens often in place names, especially longer ones.

Thus, Leicester and Gloucester were once pronounced as they’re written, with a /tʃ/ (or perhaps /k/); but that consonant was weakened over time and eventually disappeared altogether, leaving the vowels free to contract into single monophthongs, too.

  • 1
    It's interesting that the place names would derive from Roman camps (which even in temporary use tended to be very well-constructed), and today in the United States there are many cities which grew up around permanent military posts, many of which no longer feature the posts, but the names persist: Fort Wayne (Indiana); Fort Worth (Texas); Fort Collins (Colorado); and many more. Sep 19, 2014 at 15:50
  • 4
    And then there’s Legeceasterscir > Cestrescir > Cheshire > kitty-grin :).
    – tchrist
    Sep 19, 2014 at 16:10
  • Want to pop some comment on "Chichester" in? Unless some other development is already "invisible", one would have thought it should have resulted in *Cicester pronounced sister!
    – Dɑvïd
    Sep 19, 2014 at 19:44
  • 2
    @Davïd Chichester is probably from Cissa + ceaster, which is slightly different again: here we have before ceaster a non-initial, unstressed syllable (probably a schwa), which apparently got syncopated out before the weakening of the intervocalic took place. That left Cissceaster, which was then simplified to just Ciceaster, now spelt Chichester. At least that would be my guess at the chronology. (Incidentally, ‘sister’ is an attested and occasionally used contracted pronunciation, but of Cirencester.) Sep 19, 2014 at 23:08
  • 1
    @vectory Palatalisation is the standard name of the process which turned PG *k into OE /tʃ/ before front vowels. It doesn’t matter that the resulting sound is not strictly palatal; it remains the name of the process. I have no idea what you mean by “hear no trace of it today”. It’s a diachronic change, so obviously we can only hear one stage at any synchronic point in time. It is, however, one of the most basic and easily-proven sound changes that occurred in/before OE, which hundreds of examples, many even within derivations from the same roots (e.g., drink vs drench). Dec 26, 2019 at 22:59

The cester/chester ending can be traced back to the Roman occupation: castra is Latin for camp, so these placenames indicate Roman army encampments (or more likely permanent garrisons, given the names' persistence).

The Grammarphobia blog gives a good run down of the evidence which traces these pronunciations through history; the scansion in Shakespeare, for example, suggests that the elision was in common use in his time. The article also speculates on why the pronunciation and spelling are so divergent: perhaps because the people writing down the placenames (the literate scribes of Domesday Book; the monarch's tax-collectors &c) weren't the ones living there, coupled with the usual spelling free-for-all that only started to be regularised with mass-printing.

Sometimes pronunciation can swing back into line with spelling; while Cirencester is now generally pronounced as spelled, it had been elided to cicester.

Similar elisions are common in other patterns of placenames (Kirkby:kirby; Kircudbright:Kurkoobree; Marylebone:marleybone/mairbun/m'bn) and so probably represent a universal trend towards elision in familiar and locally-shared placenames.

The relative lack of shortening in such names as Manchester, Lancaster, Chichester, Colchester looks to be caused by the middle consonant 'catching' the speaker and preventing further elision. The examples in the question, by contrast, don't have a strong middle consonant.

  • Oh, dear, I knew I should have waited to see if @Janus Bahs Jacquet was going to post...
    – Flaneurben
    Sep 19, 2014 at 15:38
  • And then there's "Milngavie"!
    – Dɑvïd
    Sep 19, 2014 at 19:55
  • @Flaneurban Our answers compliment each other nicely (and I've upvoted yours)—plus I think you actually finished yours before I got mine in, by about a minute or so. (Currently on mobile device, so can't check time stamps, but I seem to recall the “A new answer has been posted” bar while typing.) Sep 20, 2014 at 12:54
  • I'll certainly be upping my game in future; currently researching IPA shortcuts
    – Flaneurben
    Sep 22, 2014 at 12:43
  • +1 Your post was good too, perhaps better in that it addresses the dropping or otherwise of the middle vowel, which after all is what freaks people out about the examples! Re IPA, I use the site I've linked below. It's great. Can use keyboard + shift of click on symbols (keyboard + Shift goes on what the symbol looks like - not sound). Can cut n paste into here or word docs etc. Really useful Type IPA Sep 25, 2014 at 10:37

I would consider my answer personal observation due to a software project. I believe it is a general trend. It would be helpful, if people would offer corrections to my observation.

Human attention span seems to constrain a concept to be pronounced no longer than 3 syllables. This human tendency is due to the angst of wanting to say as much as possible within a given time.

I can say this about English and Hebrew, since these are the languages I am familiar with. Perhaps someone could verify that French, Chinese, Hausa, Japanese, etc succumb to the same constraint.

In standard British English, three syllable words are often contracted to two-syllables during pronunciation. If that is not possible or inconvenient, then words are normally contracted to 2.5 syllables.

In standard British English pronunciation, a word usually has 2 major syllables and 1 minor syllable = 2.5 syllables. Another frequent pattern is two major and two minor syllables = 2 + 0.5 + 0.5 = 3.

These are the most common patterns of pronunciation (M=major, m = minor syllable) for three syllable words.
  1. M-m-M
  2. m-M-M
  3. m-M-m-M
  4. M---M

Frequently, pattern #2 is contracted to pattern #4.

For those rare words, or chemical names, a word is broken up into groups where the pronunciation of a group conforms to the above patterns.

There also is the principle of emphasis on the first syllable, so that a word is pronounced


in order for the pronunciation of m1 to be sufficiently diminished to allow M2 to effectively be the first syllable. That is, at all times emphasize the pronunciation of the 1st syllable - but if that sounds silly, suppress the 1st syllable to allow the 2nd Syllable the prominence of a 1st syllable. (Beyond the scope of this question, I have developed a syllable valency rating which documents what syllables get to be suppressed - such that the lower the valency, the higher its "risk" of being suppressed.)


  • Normally => M-m-M => M---M (NORM'ly)
  • Leicester => M-m-M => M---M (LE'Ster)
  • Gloucester => M-m-M => M---M (GLO'Ster)
  • Worcester => M-m-M => M---M (WOOR'Ster) => (WOO'Ster)
  • Elucidate => m-M-m-M => m-M---M (eLiUS'sdate)
  • Empowerment => m-M-m-M => m-M---M ('mPOW'rment)
  • Montgomery => m-M-m-M => m-M---M (m'GUM'mry)
  • Government => M-m-M => M---M (GOV'ment)

Compare the contraction of Government to common Indian English
Government => M-m-M => M---M (GOU'ment)


  • Worcester shire => M-m-M M-m => M---M (WOO'Ster) M---(SHier)
    => M---m-M--- (WOO'ste'SHier)

Another phenomenon is the metathesis of multiple single syllable words into one word group to be pronounced as if they were a single word conforming to the three syllable patterns.

  • Don't {Eat it up} => Don't {M-m-M EAT'iTUP}
Other forms of contractions are having
  • T pronounced as a click. e.g. Often => Of^n, Bottle => Bot^l
  • D smoothened to J (graduate => gra juate, procedure)
  • T smoothened to Ch (virtual)

OTOH, there is the reverse-contraction phenomenon, which is rare in British English but of which Americans are excessively guilty. The extension of 2 syllable words to 2.5 or 3 syllables

  • Emty => Empty (England)
  • Coupon => KiUpon (US)
  • Figure => Figear (US)
  • Nuclear => NOO-kiu-lear (GWB)
  • German agglutinates like crazy and does not follow your rule. On the other hand, what on earth does this have to do with the Question? BG, I mean no offense, but this "answer" is more a blog post than a StackExchange participation. Sep 19, 2014 at 15:55
  • 2
    I don't buy your three syllable limitation. There are many words with more than 3 syllables. The rest is quite interesting but I was more interested in the specific history of those words.
    – terdon
    Sep 19, 2014 at 15:56
  • "There are many words with more than 3 syllables" - as I stated, such words are broken into groups, where the pronunciation of each group conforms to the three syllable patterns. Sep 19, 2014 at 16:20
  • 2
    I don't think the downvotes are deserved, but putting your motivation of a software project plus a lot of general stuff was an initial turnoff. You might be better off posting your own question to get answers for the general phenomenon of contraction, and reference that with a shorter answer that's more specific to the place names here. Sep 20, 2014 at 11:10
  • @BlessedGeek: if you ever did post a question of your own, could you provide the link here in the comments or in your answer? (just upvoted)...
    – Fabby
    Feb 12, 2015 at 10:25

So yes, in Shakespeare's First Folio, you'll see the spelling "Gloster"...

...but I just figured that if you organize the letters into syllables differently, then each word is indeed spelled just as it is pronounced.




I might even speculate that "Manchester" is actually being pronounced Manche|ster, because if you try to say "Manche" followed by "ster", the reduced "e" automatically inserts itself... and generally, when I hear Brits pronounce the name, I hear that the e is indeed reduced. [EDIT: Likewise Chiche|ster.]

  • 1
    So, what you're saying is that if you keep the pronunciation as it it and break the word into syllables, then it's pronounced as it's spelled. Well, yes, obviously. That's 'cause you assume the pronunciation. For Worcestershire, for example, you seem to be pronouncing Worce as wus, and shire silent. Neither of these would be pronounced that way in other contexts so I don't see what point you're making.
    – terdon
    Sep 26, 2014 at 11:20

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.