I recently noticed a pattern regarding words ending in the /i.ə/ sound sequence:

• When the sequence is spelled as “ea,” the emphasis is on the syllable that contains the E. Examples include the words “Nicea” and “IKEA.”
• When the sequence is spelled as “ia,” the emphasis is on the syllable before the one that contains the I. Examples include “Syria,” “Serbia,” “Bosnia,” and “logia.” This even happens in -ia ending words that aren’t pronounced /i.ə/, such as “Russia,” “Croatia,” and “Micronesia.”

What is the cause of this pattern? I should also note that I found two exceptions: “Maria” (though if I remember correctly, the way that this is pronounced in Attack on Titan is not an exception) and “sharia.” Primal Groudon (talk) 17:58, 13 March 2025 (UTC)[reply]

Without addressing the central question, I note that your 'ia' examples are all 'natural language' words/names, while your 'ea' ones are a variant spelling (of Nicaea) and a modern acronym. You are also ignoring the languages of origin of the words/names (often not English), and variations in pronunciation (I have frequently heard "SHAria" in the UK, whether or not it be 'wrong'). I think you need a deeper analysis of a wider Speech corpus. {The poster formerly known as 87.81.230.195} 94.2.64.108 (talk) 18:14, 13 March 2025 (UTC)[reply]

"Area", "nausea", Echinacea? --20:23, 13 March 2025 (UTC)

Also azalea and cornea.  ​‑‑Lambiam 22:01, 13 March 2025 (UTC)[reply]

Idk whether or not to be embarrassed that I’ve now been presented with more exceptions for the ending that I couldn’t think of exceptions for than the one I could, and that the exceptions included relatively common words. Didn’t know about that third one, though. Primal Groudon (talk) 01:37, 14 March 2025 (UTC)[reply]

If the word was borrowed from Classical Latin, the word stress is almost always the same as in Latin. The ⟨i⟩ that is the penultimate syllable of Syria is short in Latin: Sȳrĭa, which, according to the stress rule for Latin, means the stress is on the syllable before it, the antepenultimate syllable. This is the most common stress pattern for Latin words ending on ia, but there are exceptions; for example, the ⟨i⟩ in the Biblical name Abīa (Abijah in English) is long and therefore gets the stress. Among the words ending on -ea borrowed from Classical Latin, many end on -aea or -oea in Latin. In these cases, the penultimate syllable ⟨ae⟩ or ⟨oe⟩ is long and gets the word stress.  ​‑‑Lambiam 22:30, 13 March 2025 (UTC)[reply]

I found new examples that aren’t proper nouns: phobia (and by extension every word that ends with -phobia) and hernia. Primal Groudon (talk) 13:59, 14 March 2025 (UTC)[reply]

One more -ia word with stress on the ⟨i⟩: peripatia.^[1] This ⟨i⟩ stems from the Greek diphthong ει in περιπέτεια, which is long. It is somewhat exceptional; English words derived from Greek words ending on -εια generally end on -y (for example, Ancient Greek ἀπάθεια → Latin apathīa → French apathie → apathy).  ​‑‑Lambiam 08:03, 15 March 2025 (UTC)[reply]

Seems as if peripetia and peripety both exist, although the most common form is peripeteia. 惑乱 Wakuran (talk) 12:38, 28 March 2025 (UTC)[reply]