As many readers realize, the pronunciation of many English consonants change based upon whether or not the preceding consonant was voiced. For example, consider the word “cards”. The pronunciation of z and s is the same, except for the fact that z is voiced and s is unvoiced. The same can be said for d and t, g and k, v and f, as well as several other pairs. The s in “cards” is pronounced as a z because the preceding d is a voiced sound.

An “s” at the end of a word is pronounced like a “z” when it follows a voiced sound.

In “carts” on the other hand, the s follows an unvoiced letter, and so the s sounds like an s. Observe the words “beads”, “beets”, “rags”, “racks”, “stops”, and “snobs”. The rule is is apparent from these words. In fact if you tried to pronounce a word like “beads” with an s instead of a z it would become very clear how strong this rule is in English.

“But a “ce” is pronounced as an “s”, almost never as a “z”

Note: The above rule doesn’t apply to every voiced consonant. An r for example, can be followed by either s or z sounds. An example of this would be the words “peers” and “pierce”. In such situations it’s impossible to tell from the spelling alone whether an s is pronounced as an s or as a z. In contrast, ce is always pronounced as an s. I’ll leave it as an exercise for the reader to determine which voiced consonants can be followed by unvoiced consonants in the same syllable. If anyone’s really interested in the answer, just post below.

“The pronunciation of an “x” depends on whether or not the syllable is stressed. It doesn’t matter if the preceding sound is stressed or not.”

Okay, all of that is stuff I’ve known for a long time. However, I came to a new realization last month while I was tutoring my one of my boss’s daughters. There’s also a rule about when x is pronounced as a ks and when it’s pronounced as a gz. Surprisingly it is not based on whether the preceding letters are stressed or not. It’s based on which syllable is stressed. In “exercise”, “Mexico”, “mix”, “axle” and “extra” the syllable ending in x is stressed. In each of those words, the x is pronounced as a ks. In contrast, “examine”, “exalt” and “example” are stressed on a syllable starting with x, and the x is pronounced as gz in all three words.

If an “x” is at the end of an accent stress, pronounce it as “ks”.
If an “x” comes before an accent stress, pronounce it as a “gz”.

Can you think of any words with “x” that don’t follow the rule?

Update: Thanks to Mike’s comment, there’s an addition to this rule:

If a word begins with “x”, the first syllable is stressed and the “x” is pronounced as a “z” instead of the normal “gz”.