Originally, the word "tea" was pronounced "tay", which would suggest that a simple "e" is short (pronounced "eh") and by adding the "a", it becomes long "ay".

However, we also have the diphthong "ae", often found in Latin words which seems to be pronounced the same way, "ay", like a long A.

So, is this just two different diphthongs that happen to be pronounced the same way coincidentally, one coming from English and the other from Latin, or is there a subtle difference in pronunciation between the two?

  • 1
    Not sure why you'd think that. In few English words is "ea" pronounced "ay" (i.e. [ei]). Break & great are 2. But the pronunciations [i:] as in "sea" and [ɛ] as in "dead" are commoner. As for "ae", that's often pronounced [i:], too, as in algae and Caesar. Coincidence, I'd say, especially as one comes mainly from Latin as you say and the other doesn't.
    – Rosie F
    Commented May 28, 2018 at 5:07
  • 5
    These are not diphthongs. They are digraphs.
    – fdb
    Commented May 28, 2018 at 20:19

2 Answers 2


Latin æ represented /ai̯~ae̯/ and monophthongized later to /e:/. The digraph simply ea represented the sequence /ea/.

Old English æ represented the vowel /æ/ or /æ:/, and ea represented the diphthong /æɑ̯/ or /æːɑ̯/. In Middle English, /æ/ merged with /ɑ/ as /a/, and /æ:/ heightened to /ɛ:/ and was written ea.

You can read about vowels in Latin or Old English on Wikipedia.

Your observation that e is short and adding a makes it long isn't exactly correct. The long version of e /e/ in Middle English orthography was ee /e:/, and ea was a separate long /ɛ:/ (historically from Old English /æ:/).

The fact that some English words spelled with ea (such as great) are pronounced with the vowel /eɪ/ has nothing to do with the pronunciation of æ as /e:/ in later Latin.

  • Middle English also developed /ɛ:/ from open-syllable lengthening or compensatory lengthening of /e/ when final schwa was lost (I’ve seen different descriptions of the sound change) Commented May 28, 2018 at 13:31
  • I understood the part about "adding the a" to be about the orthography, not about the sources of the /ɛ:/ represented by ea
    – b a
    Commented May 28, 2018 at 13:40
  • what I meant was more that the wording “the long version of Middle English e /e/ was ee /e:/“ may be a little misleading, since past a certain point /ɛ:/ (not necessarily spelled “ea”, but a long vowel even when spelled with a single vowel letter) behaved like a “long version of /e/” in Middle English Commented May 28, 2018 at 13:42
  • ...and i’ve seen sources that suggest this sound change is evidence that ME /e/ had a quality a bit opener than IPA [e] by late ME (compare how Greek epsilon lengthening to a spurious diphthong “ei” is taken as evidence for epsilon having a closer quality than eta in Classical Greek), so I’m not sure I agree with this edit of the answer either Commented May 28, 2018 at 13:50
  • @sumelic I changed the wording again, but I think the interesting points you bring up are probably less suited as a tangent to the answer, so I'll leave it to the comments
    – b a
    Commented May 28, 2018 at 14:04

As fdb said in a comment:

These are not diphthongs. They are digraphs

It probably is not too surprising that digraphs containing the letter "e" would be used to represent sounds in the area of [e] or [ɛ].

Any further connection than that seems like mainly a coincidence to me. Both of these digraphs have complicated histories involving many sound changes over the centuries.


Old Latin had a sound spelled "ai", whose pronunciation we reconstruct as something like [ai̯] or [aj]. In Classical Latin, this sound came to be spelled as "ae", which may indicate that it had come to be pronounced as a diphthong with a final element somewhat opener than IPA cardinal [i]--something like [aɪ̯] or [ae̯]. In the Romance languages, the reflex of this sound is generally the same as the reflex of Latin "short e", whose pronunciation is reconstructed as [ɛ]—this contrasts with the reflex of Latin "long e", whose pronunciation immediately preceding the loss of vowel length is reconstructed as [eː] (after the loss of vowel length, the reflex of Latin "long e" merged with the reflex of Latin "short i" in the ancestors of most Romance languages to become *e). Based on this, it is thought that there was a progression from [ai̯] > [ae̯] > [ɛː] > [ɛ] (and then on to various other reflexes in particular Romance languages). After "ae" was monophthongized, it often came to be replaced in spelling with "e" in Medieval Latin, but the modern tradition of writing Latin has returned to using "ae", and English students who learn "Restored Pronunciation" are taught to use the English "price" vowel /aɪ/ as an approximation to the Roman pronunciation. English speakers who are using a pronunciation modeled after the "Ecclesiastical" or "Italian" tradition of pronouncing Latin may use the "face" vowel /eɪ/ for "ae", although written guides to Ecclesiastical Latin often recommend that English speakers use the "dress" vowel /ɛ/ instead. Word-final DRESS is foreign to English phonotactics, which may be one reason why even "Ecclesiastical"-influenced English speakers don't tend to use word-final /ɛ/ in the pronunciation of English words from Latin.

In the traditional English pronunciation of Latin, "ae" came to be pronounced as /iː/ because of the Great Vowel Shift, except for when it was shortened due to being in a closed syllable or a "trisyllabic laxing" context: then the reflex was /ɛ/. However, there was also a tradition (I'm unsure about how old it was) of using the "long" value /iː/ for "ae" in all positions, which has left its mark in the form of certain, chiefly British English pronunciations of some English words or names that were taken from Latin and that are spelled with "ae" (e.g. "aesthete" or "Daedalus").


As ba mentioned, the digraph "ea" was used in Old English to represent both a long diphthong whose pronunciation is reconstructed as /æːɑ̯/, and a short diphthong whose pronunciation is reconstructed as /æɑ̯/. The letter "e" was used in Old English to represent both a short vowel /e/ and a long vowel /eː/. The letter "æ" was used to represent both a short vowel /æ/ and a long vowel /æː/, and the letter "a" was used to represent both a short vowel /ɑ/ and a long vowel /ɑː/.

So if we look at the Old English origin of the digraph "ea", it's not a matter of "e" representing a "short" "eh" sound and "a" being added to make this a "long" "ay" sound. The association of those sounds with these written representations develops much later on.

By the time of Middle English, short "ea" had merged with short "æ" and short "a" to become some single low vowel phoneme like /a/, typically written "a" (Middle English sound changes would lead to the development of a distinct lengthened phoneme /aː/), and long "ea" had merged with long "æ" to become a single front mid-low vowel phoneme /ɛː/, which contrasted with mid-high /eː/, from Old English /eː/. My understanding is that Middle English continued to use "e" for both short /e/ and long /eː/, but the digraph "ee" was also used for long /eː/, as well as long /ɛː/. The digraph "ea" was also used for long /ɛː/.

There is also another important source of Middle English /ɛː/. In Middle English, a sound change occurred that affected the pronunciation of many vowels. I have seen this sound change variously described as open-syllable lengthening (in certain contexts) or as compensatory lengthening of a vowel in an open syllable when the following syllable lost a schwa sound. Whatever the exact nature of the conditioning factor, the result of the sound change was that short /e/ developed to long /ɛː/ in a number of words. These words with /ɛː/ could have the spelling pattern "eCe" or "ee", but for whatever reason, after the Middle English period the spelling pattern "eCe" became uncommon and the spelling pattern "ee" became more restricted to words with the higher vowel quality, so a number of words that originally had short /e/ in Old English ended up being spelled with the digraph "ea" in modern English. An example is "meat", from Old English mete [mete], Middle English mete [mɛːt(ə)].

The "Great Vowel Shift" that is considered the boundary between Middle English and Modern English includes the sound changes /eː/ > /iː/ and /ɛː/ > /eː/; this resulting in the digraph "ea" representing the quality /eː/. The subsequent development of this /eː/ varied between dialects: in English "RP", it merged into /iː/ in all but a handful of words (like great, steak, break); in some other dialects, it may have been regular for it to merge with the reflex of Middle English "long a" and "ay~ai", although there are also some arguments for certain dialects only having a "near-merger" of the "ea" and the "long a" sound. The merger of the "ea" sound with the "long a/ay" sound is the point where we get pronunciations like "tea = tay".

Middle English /ɛː/ was sometimes shortened instead, somewhat sporadically, which resulted in Modern English /ɛ/, as in "deaf" and "dead".

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